Molecular Neurobiology (2019) 56:5436–5455 https://doi.org/10.1007/s12035-018-1448-3

The Role of and Sphingosine-1-Phosphate in Alzheimer’s Disease and Other Neurodegenerative Disorders

Kinga Czubowicz1 & Henryk Jęśko2 & Przemysław Wencel1 & Walter J. Lukiw3 & Robert P. Strosznajder1

Received: 2 August 2018 /Accepted: 6 December 2018 /Published online: 5 January 2019 # The Author(s) 2019

Abstract Bioactive sphingolipids—ceramide, sphingosine, and their respective 1-phosphates (C1P and S1P)—are signaling molecules serving as intracellular second messengers. Moreover, S1P acts through G -coupled receptors in the plasma membrane. Accumulating evidence points to sphingolipids' engagement in brain aging and in neurodegenerative disorders such as Alzheimer’s, Parkinson’s, and Huntington’s diseases and amyotrophic lateral sclerosis. Metabolic alterations observed in the course of neurodegeneration favor ceramide-dependent pro-apoptotic signaling, while the levels of the neuroprotective S1P are reduced. These trends are observed early in the diseases’ development, suggesting causal relationship. Mechanistic evidence has shown links between altered ceramide/S1P rheostat and the production, secretion, and aggregation of amyloid β/α-synuclein as well as signaling pathways of critical importance for the pathomechanism of protein conformation diseases. Sphingolipids influence multiple aspects of Akt/protein kinase B signaling, a pathway that regulates metabolism, stress response, and Bcl-2 family . The cross-talk between sphingolipids and transcription factors including NF-κB, FOXOs, and AP-1 may be also important for immune regulation and cell survival/death. Sphingolipids regulate exosomes and other secretion mechanisms that can contribute to either the spread of neurotoxic proteins between brain cells, or their clearance. Recent discoveries also suggest the importance of intracellular and exosomal pools of small regulatory RNAs in the creation of disturbed signaling environment in the diseased brain. The identified interactions of bioactive sphingolipids urge for their evaluation as potential therapeutic targets. Moreover, the early disturbances in sphingolipid metabolism may deliver easily accessible biomarkers of neurodegen- erative disorders.

Keywords Alzheimer’sdisease . Ceramide . Huntington’sdisease . microRNA . Parkinson’sdisease . Sphingosine-1-phosphate

Introduction the development of devastating disorders. Two of the most widespread neurodegenerative diseases are Alzheimer’s(AD) Aging, which itself influences the central nervous system (CNS) and Parkinson’s (PD). Dementia affects estimated 47 million in a relatively subtle manner, creates vulnerable background for people worldwide [1], placing enormous burden on the affected individuals, their families, societies, and healthcare systems. AD is the most common neurodegenerative disorder, responsible for Kinga Czubowicz and Henryk Jęśko contributed equally to this work. up to 70% of dementia cases [1]. This disease is characterized by the presence of aggregates of pathologically misfolded proteins, * Robert P. Strosznajder including the extracellular senile plaques built mainly of amy- [email protected] loid β (Aβ), a product of proteolytic cleavage of the transmem- β β β γ 1 Laboratory of Preclinical Research and Environmental Agents, brane A precursor protein (A PP) by -and -secretase [2]. Department of Neurosurgery, M. Mossakowski Medical Research Neurons in AD also display cytoskeletal abnormalities that are Centre, Polish Academy of Sciences, 5 Pawinskiego St., linked to hyperphosphorylation and aggregation of the PL-02106 Warsaw, Poland microtubule-associated tau protein into intracellular neurofibril- 2 Department of Cellular Signalling, Mossakowski Medical Research lary tangles [3]. Centre, Polish Academy of Sciences, 5 Pawinskiego St., Parkinson’s disease is the most frequently occurring move- PL-02106 Warsaw, Poland ment disease and the second most widespread neurodegener- 3 LSU Neuroscience Center and Departments of Neurology and ative disorder after AD [4]. It is estimated that PD affects up to Ophthalmology, Louisiana State University School of Medicine, New Orleans, LA 70112, USA 1% of people over the age of 60 and up to ca. 4% over 85 [5]. Mol Neurobiol (2019) 56:5436–5455 5437

PD is characterized by subcortical neurodegeneration, includ- pathway. Sphingosine kinases (SPHK1, SPHK2) phosphorylate ing the characteristic loss of dopaminergic phenotype neurons sphingosine into S1P in a highly regulated fashion in various in substantia nigra pars compacta, loss of dopaminergic stria- cellular compartments; dephosphorylation is carried out by S1P tum innervation, and histopathological aberrations in the form phosphatases (SGPP1 and SGPP2), while S1P can be also hy- of α-synuclein (ASN)-containing intracellular Lewy bodies drolyzed irreversibly into ethanolamine phosphate and (LB)/Lewy neurites (LN) [4]. Disturbances in other neuro- hexadecenal by S1P lyase (SGPL) [9]. Activity of the enzyme transmitter systems (serotoninergic, noradrenergic, and cho- glucocerebrosidase (GBA) can be another ceramide source [30] linergic) are increasingly being recognized along non-motor with significant links to Parkinson’sdisease(seePt.'Theroleof symptoms, which—in later stages—may include dementia bioactive sphingolipids in Parkinson’s disease'). [6]. Like AD, Parkinsonian neurodegeneration progresses in a stealthy manner, and when clear symptoms appear the do- S1P and Ceramide in Neuronal Survival and Death paminergic neuron population is already decimated [4]. Signaling Bioactive Sphingolipids Biosynthesis Bioactive sphingolipids employ several mechanisms to exert their influence on intra- and extracellular signaling pathways. Once merely considered structural compounds, bioactive Plasma membrane is one of main cellular regions of SPHK sphingolipids are increasingly implicated as signaling mole- activation; S1P and probably C1P can bind cell surface receptors cules in the brain, and play important roles in aging, neurode- [31, 32]. S1P receptors (S1PR1 to S1PR5) belong to the Edg generative disorders, and the accompanying immune deregula- family and bind G ,G,G , and Rho proteins [31]. While tion [7]. Ceramide, ceramide-1-phosphate (C1P), sphingosine, q i 12/13 data on the postulated C1P receptor is scarce, all S1PR1 to -5 and sphingosine-1-phosphate (S1P) are the best described bio- bind G ,S1PR2and-3bindG and all but S1PR1 signal through active sphingolipids regulating stress resistance, proliferation, i q G (Fig. 1). Through either S1PR-binding G proteins or differentiation, and mature phenotypes of nervous system cells 12/13 through more direct interactions with intracellular enzymes, [8, 9]. Sphingolipids have multiple ancillary roles in the regu- sphingolipids also modulate signaling events such as cAMP lation of cell growth, death, senescence, adhesion, migration, (cyclic adenosine monophosphate), MAPK (mitogen-activated inflammation, angiogenesis, and intracellular trafficking in the protein kinase), PKC (protein kinase C), PLD (phospholipase CNS [10, 11]. The sphingolipid rheostat model ascribed these D), and PI3 kinase–Akt–PKB (protein kinase B) [35, 36]. The compounds clearly opposite roles in cellular survival signaling: PI3K–Akt pathway is a crucial integrator of metabolic and stress ceramide as a cell death activator, while C1P and S1P promoted signals engaged in a plethora of physiological and pathological survival. The fact that single phosphorylation reaction turns processes ranging from energy metabolism through aging to ceramide and sphingosine into their antagonistic counterparts age-related diseases [8]. Sphingolipids are also structural com- stresses the significance of the precise, successful control of ponents of lipid bilayers. Changes in their local concentrations sphingolipid metabolism enzymes [8, 12]. Although the pro- modify membrane properties further modulating signaling versus anti-survival roles have blurred somewhat in recent years events that take part in these membranes. [13], the critical roles of sphingolipid signaling in nervous sys- tem function have been confirmed by effects of mutations in CerK C1P C1PR? C1PP their biosynthesis and receptor [14–24]. The importance Cer PKC, CerS Ceramidase 2+ of bioactive sphingolipids is also stressed by the accumulating S1PR2, S1PR3 Gq PLC, Ca cNOS Sph evidence about their involvement in aging and neurodegenera- SphK FOXOs, SGPP PI3K, Akt – GSK-3β, tive disorders [8, 25 29]. S1P S1PR1-S1PR5 Gi Therateofceramidebiosynthesis is controlled by the first step ERK Bax, Bad of the pathway’s de novo branch, which is catalyzed by serine S1PR2-S1PR5 G12/13 Rho NF-κB palmitoyltransferase (SPT). SPT product dihydrosphingosine is then metabolized by ceramide synthase (CERS) to Fig. 1 Signaling pathways triggered by the cell surface receptors for dihydroceramide, which is subsequently converted by S1P (S1PRs) and C1P (C1PR). S1P through its G protein-binding dihydroceramide desaturase (DES or DEGS) to ceramide. receptors modulates pathways known for their engagement in the Ceramide may be metabolized into sphingomyelin by regulation of cellular metabolism oxidative/nitrosative stress and death/ survival. The depicted examples of S1PR-activated signaling pathways sphingomyelin synthase (SMS, or SGMS); the reverse reaction are far from exhausting the spectrum of observed interactions (e.g., (the sphingomyelinase pathway) catalyzed by S1PR2 inhibits Akt while S1PR3 activates it; iNOS is typically induced sphingomyelinases (SMases, SMPDs) is another major ceramide by NF-κB, but can also undergo S1P-/C1P-dependent suppression via source. Ceramide also serves as a precursor for the production of p38 MAPK [8, 33]). In contrast, C1PR, the cell surface receptor for ceramide-1-phosphate, remains poorly characterized and it is generally sphingosine by ceramidases. CERS can perform the opposite not known what part of C1P effects it mediates. According to [34], reaction, which is third ceramide source, termed the salvage modified 5438 Mol Neurobiol (2019) 56:5436–5455

Ceramide controls not only multiple cell death mechanisms 3b). ERK appears to mediate the pro-survival action of S1P [69]. but also cellular senescence, differentiation, and aspects of S1PRs also stimulate the anti-apoptotic PI3K-Akt pathway [70], arborization in neurons [37, 38]. Sphingosine also seems to whose disruption in AD may heavily contribute to the disease be engaged in cell death modulation [39]. C1P has been pathomechanism [71, 72]. The nuclear transcription factors shown to stimulate cellular survival, growth, and may coun- targeted by S1P-sensitive pathways include FOXO3a (inhibited teract ceramide signaling also through downregulation of acid by the PI3K-Akt pathway [70]), AP-1 (a transcription factor sphingomyelinase and serine palmitoyltransferase activities receiving input from Jnk/p38/ERK [37, 73] and engaged in the [40, 41]. S1P regulates cell viability, neuronal excitability, network of mutual co-regulation between sphingolipid-related and arborization [31]. Sphingolipids are also engaged in im- genes [48–50]), or NF-κB (nuclear factor κB, through direct mune phenomena, which critically alter the fate of brain cells interaction between SPHK1 with TNF receptor-associated fac- in neurodegenerative disorders [31, 37, 38, 42, 43]. tor TRAF2, and through S1P acting as TRAF2 cofactor [74, The roles of S1P and ceramide in the survival of brain neurons 75]). Moreover, histone deacetylases (HDAC1 and -2) are are far more complex than the antagonism described in the inhibited through S1P binding [59]andcanblockNF-κBvia sphingolipid rheostat model (Fig. 2) and dependent on the sig- its deacetylation [76]. NF-κB influence on cell death may vary naling milieu (see below). However, it is highly probable that the depending on the signaling context, immune activation, etc. AD-linked changes in the metabolism of bioactive sphingolipids Through PI3K-Akt, S1PRs can inhibit GSK-3β (the cru- should significantly alter the rates of neurodegeneration. cial tau kinase [77]) and the pro-apoptotic protein Bad. In In most situations, S1P and ceramide antagonistically signal addition, S1P has been shown to inhibit ceramide production cell survival or death largely via shared mediators (Fig. 3a, b). by acid sphingomyelinase (aSMase) [63]. However, contrary Yet more strikingly, both can directly inhibit each other’ssynthe- to the initial view on the clear-cut S1P-vs.-ceramide opposi- sis [63, 64]. S1P is classically viewed as an anti-apoptotic agent tion, in some situations, S1P may actually exert neurotoxic [65, 66] and has been shown to mediate the actions of numerous influence—depending on the spatiotemporal control of its anti-apoptotic compounds (Fig. 3b, reviewed in [67]). S1P typi- production and degradation, or when its concentration reaches cally opposes the pro-apoptotic role of ceramide presumably by too high levels [78]. Moreover, it is necessary to bear in mind decreasing oxidative stress and modulating the expressions of the mentioned ambiguous nature of some of S1P’smediators: pro- and anti-apoptotic proteins of Bcl-2 family (Fig. 3)[47]. AP-1 [79], ERK [80, 81], or NF-κB[82, 83]. S1P can activate p38, ERK, and block Jnk in various tissues The roles of in cellular homoeostasis reach far by acting through its surface G protein-coupled receptors [67] beyond just being pro-apoptotic molecules. Loss of physiolog- (however, S1P influence on Jnk can be more varied [68]) (Fig. ical ceramide concentrations can lead to structural disturbances

Fig. 2 The changes in bioactive sphingolipid levels observed in aging Parkinson’s Disease'). Much fewer works address the problem of and neurodegenerative disorders. Numerous observations in physiological brain aging, where the changes appear to be partially postmortem brain tissues point to the imbalance in the ratio between the gender-specific [44].The asterisk indicates clinical data about S1P in concentrations of the apoptosis inducer ceramide, and the typically anti- PD is missing; however, in experimental disease models, reduced apoptotic S1P (see Pts. 'Bioactive Sphingolipids in the Pathomechanism SPHK activity leading to loss of Akt signaling was observed [45, 46]. of Alzheimer’s Disease' and 'The Role of Bioactive Sphingolipids in Mol Neurobiol (2019) 56:5436–5455 5439

a b High ROS SPT CERS p53 Moderate DEGS ROS SMase C1PP aSMase SphKs

Cer SphK1 S1P TRAF2

PP2A ROS PI3K S1PRs

Akt Jnkp38 ERK Akt Jnkp38 ERK Bax, Bad, GSK-3β p53 AP-1 Bad, HDAC1, FOXOs NF-κB GSK-3β AP-1 HDAC2 MRC inhibition, ROS, AIF, cytochrome c, caspase-2, -3, -5, -8, -9, beclin1, LC3-II

Axonal Neuronal Inhibition of neuronal death degeneration death

Fig. 3 The role of ceramide and S1P in neurodegeneration pathways. autophagosomal LC3-II (which binds mitochondrial ceramide to induce a Ceramide-induced axon loss and neuronal apoptosis. According to [37], lethal mitophagy) [56–58]. While inhibition of HDAC1 and -2 is engaged modified. b The roles of S1P and SPHKs in the modulation of neuronal in the pro-survival signaling of S1P, the role of HDAC3 is more death. According to various authors (see text). Both S1P and ceramide(s) ambiguous [59–62]. AIF, apoptosis-inducing factor; AP-1, activator exert major part of their opposing influence on cell survival through protein-1; aSMase, acid SMase; LC3-II, lipidated microtubule- multi-level modulation of the PI3K–Akt pathway, which integrates associated protein 1 light chain 3β; C1P, ceramide-1-phosphate; C1PP, sphingolipid-based signals with clues on the metabolic condition of the C1P phosphatase; CERS, ceramide synthase; DEGS, dihydroceramide cell, stress levels, etc. [8, 37, 47]. Moreover, sphingolipid signaling desaturase; ERK, extracellular signal-regulated kinase; FOXO, forkhead displays links with the transcription factors AP-1 and NF-κB[48–52], box protein O; GSK-3β, glycogen synthase kinase 3β;HDAC,histone which regulate a plethora of processes including cell death and deacetylase; Jnk, c-Jun N-terminal kinase; MRC, mitochondrial inflammation. Akt targets FOXO1a, 3a, 4, and 6 are engaged in cell respiratory chain; NF-κB, nuclear factor κB; PI3K, phosphoinositide 3- death regulation in human tissues [53, 54]. The prevailing role of kinase; PP2A, protein phosphatase 2A; ROS, reactive oxygen species; elevated ceramide in cell degeneration/death is mediated by multiple S1P, sphingosine-1-phosphate; S1PR, S1P receptors; SMAC, second signals: inhibition of mitochondrial respiration and increased production mitochondria-derived activator of caspases; SMase, sphingomyelinase; of reactive oxygen species [55]; the release of AIF, cytochrome c,or SPHK, sphingosine kinase; SPT, serine palmitoyltransferase; TRAF2, SMAC from mitochondria; the Bcl-2-binding protein beclin1; TNF receptor-associated factor 2 in mitochondria and reduced respiration [84]. Ceramides also [89]. Other mitochondrial mediators of apoptosis known to be regulate membrane dynamics, thus influencing other aspects of released in neurons by ceramides include apoptosis-inducing organellar function and life cycle such as mitochondrial fusion factor (AIF), the second mitochondrion-derived activator of and fission, or vesicular transport [85]. It is also well understood caspase (Smac), and the stress-regulated endoprotease Omi that their (patho)physiological roles are highly dependent on [56]. Ceramide-induced apoptosis thus involves both caspase- their chain length/CERS isoforms [86, 87]. When signaling mediated and caspase-independent pathways. apoptosis, ceramides appear to use a spectrum of mediators As discussed above, changes in the balance between S1P largely shared with S1P, albeit often in a contrasting way (Fig. and ceramide (Figs. 2 and 3) may not only influence apoptosis 3a). Ceramides lead to dephosphorylation and inactivation of but also change the regulation of autophagy by the complex Akt via the protein phosphatase PP2A; this relieves Akt’sin- interplay between mTOR, beclin, and Bcl-2. S1P-dependent hibitory influence on Bad and GSK-3β [77]. On the other hand, autophagy is thought to be a homeostatic, pro-survival re- ceramide-associated increase in reactive oxygen species (ROS) sponse involved in the clearance of intracellular debris (dam- leads to the activation of p38 and Jnk and to ERK inhibition aged proteins/dysfunctional organelles) [67]. In AD, autoph- [37]. The combined influence of p38, Jnk, and ERK modifies agy can play critical role in the defense against oxidatively the activities of p53 and AP-1 (c-Fos, c-Jun) transcription fac- damaged cellular components, and its disturbances may exac- tors [88]. Together with Bad and GSK-3β activation, these erbate Aβ and tau deposition [90, 91]. However, autophagy changes cause mitochondrial alterations and via cytochrome c can also constitute a mode of cell death, where release and the activities of caspase-2,-3,-5,-8,and-9may autophagolysosomal degradation of mitochondria is depen- lead to axonal degeneration or neuron death [37]. Ceramides dent on the interaction between ceramide and LC3-II might also directly form pores in the outer mitochondrial mem- (lipidated microtubule-associated protein 1 light chain 3β) brane, leading to the release of cytochrome c and other proteins present on lysosomes [57](Fig.3). 5440 Mol Neurobiol (2019) 56:5436–5455

Bioactive Sphingolipids in Aging dysfunction, glucose hypometabolism/other metabolic stress- es, Ca2+ deregulation, and inflammatory response. Many of Bioactive sphingolipids have been investigated in the course these pathways are triggered and propagated due to the actions of aging and in association with extreme longevity [27, 92, of soluble oligomers of Aβ peptide on neurons and glia. The 93]. Centenarians display altered fatty acid pattern in role of ceramide/S1P was analyzed in the context of these ceramides and glucosylceramides—higher levels of damage pathways as well as the process of amyloidogenesis. sphingolipid species possibly linked to stress resistance (low oxidation susceptibility due to unsaturated fatty acid content) The Interactions Between Ceramide/S1P and AβPP/Aβ [27], while increased concentration of sphingomyelins (cer- Metabolism amide precursors) has been observed during aging [94]. Sphingolipids appear to have significant influence on the Structural roles of sphingolipids in cellular membranes includ- course of aging; research on lower organism models suggests ing lipid rafts constitute an important aspect of their engage- links between ceramide synthesis and longevity [8, 25, ment in AβPP/Aβ metabolism [128]. Lipid rafts are 95–97]. Importantly, bioactive sphingolipids are capable of cholesterol- and sphingolipid-enriched microdomains of the influencing the IIS (insulin/insulin-like signaling)–PI3K– plasma membrane described as signaling platforms [129, Akt, a highly conserved, versatile modulator of metabolism, 130]. Rafts are strongly associated with Aβ production, and aging, and stress response (Fig. 3)[98–102]. IGF-I signaling both β- and γ-secretases are enriched in these structures [129, in the brain has been identified to negatively influence organ- 131, 132]. Lipid rafts also seem to influence Aβ aggregation ism longevity also in mammals [103, 104], although some [133]. In turn oligomeric Aβ42 associates with rafts [134]; Aβ controversies persist [105]. Results obtained in humans appear can change membrane fluidity, which may exert a feedback to support IIS role in aging [106, 107]. IIS seems to redirect influence on its own production [135]. the vital resources away from long-term investment in favor of Rafts are sensitive to fluctuations in sphingolipid levels, more current needs such as metabolic regulation and cellular leading, e.g., to changed properties of membrane-associated survival. This leads somewhat surprisingly to the trophic in- enzymes or receptors. Sphingolipid/ceramide deficiency leads fluence of IIS in the brain [8, 108–111]. PI3K-Akt signaling to increased secretion of sAβPPα, the product of non- regulates SPHKs and S1PRs expression/activity and intracel- amyloidogenic cleavage. However, it also leads to enhanced lular sphingolipid transport [112–115]. In turn, S1P receptors secretion of Aβ42 possibly through modulation of raft- can differentially modulate Akt activity [116–118]. Ceramide associated proteins and changes in raft membrane properties leads to inhibition of Akt-dependent pro-survival signaling resulting in altered α-vs.β-cleavage ratio [136]. Exogenous [47, 119, 120], while C1P stimulates it [121, 122]. addition of ceramide and elevated endogenous ceramide in- The links between sphingolipids and cellular stress are an creased the level of Aβ. C6-ceramide, a cell-permeable cer- extremely important aspect of their potential involvement in amide analogue, increased the rate of Aβ biosynthesis by aging (Fig. 3)[8 , 67]. SPHK1 might inhibit ROS and reduce affecting β-cleavage of AβPP. Lipid raft ceramides stabilize sensitivity to DNA damage [123]. S1P and ceramides are BACE1 (β-site AβPP cleaving enzyme 1, a β-secretase) under positive influence of the stress sensor p53, and faulty [137]. Additionally, it was shown that synthetic ceramide an- ROS control leads to alterations in S1P/ceramide signaling alogues may also function as γ-secretase modulators that in-

[67, 124–126]. Even more than in aging, stress and inappro- crease Aβ42 production [138]. FTY720 in turn has been dem- priate stress responses are central elements of the onstrated to reduce hippocampal neuron damage and the pathomechanism of neurodegenerative disorders. resulting learning and memory deficits in a rat model induced

by bilateral, stereotactic injection of pre-aggregated Aβ42 into the hippocampus [139]. Some of the neuroprotective effect Bioactive Sphingolipids might be ascribed to mobilization of extrasynaptic, N- in the Pathomechanism of Alzheimer’s methyl-D-aspartate receptors to the synapse (a phenomenon Disease reducing cellular sensitivity to Aβ-induced neurotoxic calci- um influx) [140]. Importantly, FTY720 and KRP203 (another The pathogenesis of AD is not yet fully elucidated, and the SPHK2 substrate that can bind S1PR upon phosphorylation) actual roles of many of the observed disturbances are not clear. have been shown to reduce neuronal Aβ generation [141]. Accumulating evidence points to the involvement of bioactive However, the relationship between S1P and AβPP metabo- sphingolipids in AD starting from the earliest, prodromal lism is still obscure, as the compounds used may as well stages [127]. downregulate S1PR-dependent signaling; moreover,

Well-documented mechanisms that induce neuronal and FTY720 increased Aβ42 in mice in addition to reduction in synaptic degeneration in AD brain include the following: ox- Aβ40 [141]. Positive correlation between S1P production by idative damage, altered redox signaling, mitochondrial SPHK2 and AβPP processing has been reported [142]. S1P Mol Neurobiol (2019) 56:5436–5455 5441 produced by SPHK2 may activate BACE1, thus leading to induced by Aβ. Activation of neutral sphingomyelinase higher release of Aβ peptides. S1P was shown to specifically (nSMase) and increase in the level of ceramides was observed. bind to full-length BACE1 and to increase its proteolytic ac- Moreover, it was also shown that suppression of ceramidase tivity. The production of Aβ peptides can be reduced in N2a activity additionally increased the toxicity of Aβ [152]. In the neuroblastoma cells by pharmacological inhibition of sphin- studies of Jana and Pahan [153] superoxide-mediated activa- gosine kinases, homozygous SPHK2 deletion, or over- tion of nSMase was observed in primary culture of human expression of the S1P lyase gene and SGPP1 phosphatase neurons treated with Aβ1–42. NAPDH oxidase mediated the [142]. A shift in SPHK2 subcellular distribution from cytosol effect, because gene silencing for the p22phox subunit by to the nucleus was observed to correlate with Aβ deposition in antisense oligonucleotides inhibited the apoptosis of neurons

AD brains by Dominguez and collaborators [143]. In turn, Aβ induced by Aβ1–42. The authors showed that the use of N- production correlates with low SPHK1 and high S1P lyase acetylcysteine and the NADPH oxidase inhibitor prevented protein [144]. Disturbances in S1P observed in AD may not the generation of ceramides and protected against neuron ap- only critically regulate caspase-mediated AβPP cleavage. S1P optosis. Similar results were obtained in primary rat cortical regulates lysosomal AβPP metabolism in a calcium- neurons treated with Aβ oligomers where an increase in neu- dependent manner [145]. S1P is also a pro-secretory mole- tral and acid sphingomyelinase activities was observed [63]. cule, and the dependence of AβPP secretion on S1P signaling In another study, Gomez-Brouchet et al. [154] indicated that has direct potential significance in AD [146]. The regulation exposure to Aβ25–35 induced strong SPHK1 inhibition and of gene expression via S1PRs and through intracellular sig- ceramide accumulation in neuronal SH-SY5Y cells. In that naling can also lead to complex changes in cellular metabo- study, the cell death was prevented by overexpression of lism. AβPP modulates this process, as shown in FTY720- sphingosine kinase, whereas downregulation of the enzyme treated mice overexpressing mutant (V717I) AβPP. by RNA interference enhanced the cell death. Inhibitors of FTY720, which raises significant hopes as a repurposed neu- both SMases and exogenously administered S1P demonstrat- roprotective drug in AD, increased the gene expression of ed cytoprotection in the model. Ayasolla et al. [155]observed sphingosine kinases (SPHKs), ceramide kinase (CERK), and that in the primary culture of rat astrocytes treated with the anti-apoptotic Bcl-2 in an age-dependent manner [147]. TNF-α/IL-1β,followedbyAβ25–35, there is a greater increase The milieu of the brain affected by AD provides multiple in the expression of induced nitric oxide synthase (iNOS) and stressors such as ROS and cytokines, which could in turn lead nitric oxide production (NO) than in TNF-α/IL-1β-only as- to increased ceramide production (Figs. 3 and 4). Lee et al. trocytes. In a glial cell line treated with Aβ25–35 and LPS/ [152] showed ceramide-dependent death of oligodendrocytes IFNγ, an increase in iNOS expression and NO production

Fig. 4 Changes in brain sphingolipid metabolism and signaling protein, and SPHK protein/activities were observed along elevated S1P observed in Alzheimer’s disease. Human postmortem brain material lyase proteins in selected cortical areas and in the hippocampus [65, 144]. was used by numerous authors to compare the levels of bioactive Increased ceramide (Cer) and sphingomyelin (SM) levels, mRNAs for sphingolipids, mRNAs, proteins, and enzyme activities. In the ceramide synthases (CERS1, 2), SGPL1, SPTLC2, aSMase, and hippocampus S1P levels, mRNAs for CERK, S1PR1, SPHK1, SPHK2, sphingomyelinase protein and activity were observed in brain cortical and SPHK activity are reduced [65, 147, 148]. Lower S1P levels, S1PR1 areas, while ASAH1, CERK,andCERS6 mRNAs were reduced [148–151] 5442 Mol Neurobiol (2019) 56:5436–5455 was observed. These phenomena were accompanied by an [142] who reported upregulation of SPHK2 activity in AD increase in the level of ceramides as a result of the activation brain cortex while other authors reported reduction of its ac- of nSMase. Similar results were obtained in oligodendrocytes tivity and mRNA in the hippocampus [65, 147]. treated with Aβ25–35 and TNF-α, or with C2-ceramide and Importantly, SPHKs’ roles include engagement in the regula- TNF-α [156]. TNF-α-induced ceramide production was also tion of inflammation, a phenomenon already exploited in the observed by Martinez et al. [157]. therapy of relapsing remitting multiple sclerosis [164]. The The accumulated oligomerized Aβ peptide in AD brain known engagement of sphingolipids in the modulation of may also promote ceramide formation, as demonstrated both NF-κB signaling by TNF-α [165] and other factors strongly sug- in cell culture [148, 153–156] and animal models [158]. gests widespread opportunities in this field. Aβ specifically mod- Imbalance in mRNA expression of enzymes responsible for ulates the expression of some S1P cell surface receptors in mono- S1P to ceramide ratio, which potentially might decide of the cytes [166]. Sphingolipid modulators inhibit the accumulation of brain cell fates, is observed from the earliest clinically recog- mononuclear phagocytes in response to Aβ, leading to proposals nizable AD stages (Fig. 4)[149]. Sphingomyelin hydrolysis of their use as therapeutic agents [166]. In turn, anti-ceramide stimulated by Aβ appears to be the main source of ceramides immunity might also contribute to the disease progression [167]. in the pathology of Alzheimer’sdisease[155, 159, 160], along An important hint about the significance of sphingolipids in with de novo synthesis (expression of the de novo enzymes AD is the association of apolipoprotein E (ApoE whose poly- increases gradually during AD progression [149]). Activation morphisms are strongly linked to AD risk [168]), with the of SPT by Aβ peptides has been observed, resulting in neu- receptor-mediated signaling of secreted S1P [169]. Moreover, rotoxic increase of ceramide levels via de novo pathway [148, correlation has been observed between SPHK activities/S1P 161]. Aβ induces apoptosis by activating aSMase and content and ApoE allele (2.5× higher S1P/sphingosine ratio in nSMase, thereby contributing to the increase in ceramide the hippocampus of ApoE2 vs. ApoE4 carriers) in AD [65]. levels. Senile plaques contain aSMase and nSMase proteins Sphingolipids might be useful, accessible AD biomarkers along with high levels of saturated ceramides [63, 162], and [170–172]and—potentially—therapeutic targets [173]. aSMase activity is upregulated in human AD brains (Fig. 4) [150]. Examples of genes upregulated by AD also included S1P/Ceramide and the Exosome-Mediated Spread ceramide synthases CERS1 and CERS2, S1P lyase SGPL1,or of AD Pathology serine palmitoyltransferase catalytic subunit SPTLC2,while the acid ceramidase ASAH1, ceramide kinase CERK,or— Exosomes are sphingomyelin- and ceramide-enriched vesicles less obviously—CERS6 were reduced [131, 149]. However, created inside the multivesicular endosomes (MVE) and then contrary to the abovementioned results, Couttas et al. [28] secreted when MVE membrane fuses with the plasmalemma. have found an early loss of CERS2 activity at Braak stage Exosomes are engaged in intercellular communication and I/II (temporal cortex) to III/IV (frontal cortex, hippocampus). carry microRNAs (miRNAs), messenger RNAs (mRNAs), An interesting but underexplored link has been identified be- and protein- and lipid-based signaling molecules. Vesicles re- tween the still obscure physiological Aβ PP role and leased by Aβ-treated astrocytes contain the pro-apoptotic sphingolipid metabolism, as AβPP intracellular domain is ca- prostate apoptosis response 4 (PAR-4) protein and cause apo- pable of reducing the expression of SPTLC2, potentially keep- ptosis in naive cultures [174]. Rodent exosomes can contain ing the whole sphingolipid metabolism under negative control Aβ, BACE1, and presenilins 1 and 2 [175]. Amyloid plaques [131]. Aβ also disturbs S1P signaling, potentially shifting the in the AD brain contain an exosome marker [176]. These balance towards a much more pro-apoptotic state (Figs. 2 and results have led to a hypothesis that exosomes might seed 4)[148]. Aβ can downregulate the genes for SPHKs and Aβ aggregation [177]. However, at least under some circum- diminish the level of S1P as observed in wild-type and stances exosomes can also inhibit Aβ oligomerization and AβPP-overexpressing PC12 cells in culture [163]. The study promote its microglia-mediated clearance [178]. These results by Couttas et al. [65] showed reduced S1P levels with increas- might explain the observed association of exosomes with Aβ ing Braak stage in tissue samples taken from the CA1 region as a physiological, neuroprotective phenomenon [179], at of the hippocampus, or gray and of the inferior least in the healthy tissue. It is also possible that exosomes temporal gyrus (Fig. 4). AD brains also display upregulated of various origin (e.g., neuronal vs. astrocyte) might exert expression of S1P lyase SGPL1 and S1P-metabolizing phos- opposite influence or that the exosomal membranes might phatases [65, 149]. The studies of Ceccom et al. [144]showed facilitate Aβ aggregation independently of protein-mediated a decrease in immunoreactivity of SPHK1, S1P receptor 1, exosomal functions (e.g., Aβ degradationbyexosomal and an increase in S1P lyase in samples taken from the frontal insulin-degrading enzyme or neprilysin)—reviewed in [177]. and entorhinal cortices from human AD brains. However, the Additionally, exosomes can serve as a vehicle for the extra- complex influence of SPHK2 signaling on cell fate and neu- cellular secretion and cell-to-cell transport of ASN and tau rodegeneration is reflected by the study of Takasugi et al. protein, potentially further supporting the spread of Mol Neurobiol (2019) 56:5436–5455 5443 aggregation pathology [180, 181]. S1P receptor signaling has functionally “addressed” for different destinations, allowing been implicated in exosomal cargo sorting: activity of the passage of ASN (also oligomeric) into various compartments S1PR-regulated Rho family GTPases was necessary for the of recipient cells [191]. This may have high significance for the process, and Gβγ inhibitor blocked it [182]. Exosome secre- postulated spread of ASN-induced pathology along anatomical tion can be modulated by the activity of neutral connections [4]. sphingomyelinase 2 (nSMase2) and sphingomyelin synthase Recent evidence suggests links between sphingolipids and 2 (SMS2), suggesting unique roles for these enzymes in AD PD although data is relatively less abundant. PD is associated [178, 183], and additional significance for the disturbed cer- with disturbances in sphingolipid metabolism (Figs. 2 and 5) amide levels observed in the course of the disease, as [192, 195, 196, reviewed in 12]. Lipidomic analysis has shown discussed above. that the levels of ceramides and sphingomyelins were altered in postmortem PD brain tissue as compared to the control samples (a tendency towards shorter acyl chain) [192]. Findings in body The Role of Bioactive Sphingolipids fluids suggest the diagnostic value of sphingolipids in PD [197]. in Parkinson’sDisease In blood plasma, several saturated ceramides and one unsaturat- ed species were significantly higher in PD [198]. Additionally, The selective, spatially progressing neurodegeneration ob- some blood ceramide species were higher in PD with dementia served in PD defies full explanation, although hypotheses have than in non-demented PD patients and the levels of several been created that probably successfully identify and describe saturated ceramides associated with PD-linked psychiatric com- important aspects of its mechanism [4]. Pathological aggrega- plications [198, 199]. Mutations in the SMPD1 gene have been tion of ASN inside neuronal cells is widely associated with PD; repeatedly confirmed to correlate with PD risk [200–204]. The ASN might also play some role in AD [184]. ASN binds lipid expression and activity of SPHK1 are reduced in MPTP (1- rafts, and negatively regulates S1PR1 signaling there [130]. methyl-4-phenyl-1,2,3,6-tetrahydropyridine)-induced murine Moreover, the relatively recently recognized phenomenon of model of PD (Fig. 5)[194], which can lead to enhanced ROS ASN secretion suggests links with sphingolipid signaling, as as well as BAX and HRK (harakiri) mRNA expression [205]. the engagement of sphingolipids in neuronal secretory path- Importantly, the possible similarities between AD- and PD- ways is well documented [7, 146]. ASN may undergo regulated linked disturbances of sphingolipid metabolism are not limited secretion in a number of partially characterized mechanisms to direct regulation of apoptotic signaling; products of S1P deg- [45, 185–190], possibly leading to the peptide being radation have been found to modulate autophagic/lysosomal

Fig. 5 Changes in brain sphingolipid metabolism/signaling observed the disease, although reduction in total levels of ceramide and in Parkinson’s disease and its experimental models. Ceramide sphingomyelin concentration was observed [192]. Reduced sphingosine synthase (CERS) mRNAs, ceramide (Cer), and sphingomyelin (SM) kinase-1 and -2 (SPHKs) and S1P receptor 1 (S1PR1) expression was species were measured in postmortem PD brains by Abbott et al. observed in the mouse MPTP (1-methyl-4-phenyl-1,2,3,6- Increased CERS1 expression was found along with a shift towards tetrahydropyridine)-induced model [46, 193]. The MPTP model also shorter ceramide acyl chain lengths in brain regions most affected by displayed lower protein levels and activity of SPHK1 [193, 194] 5444 Mol Neurobiol (2019) 56:5436–5455 degradation of both Aβ/AβPP and ASN [206]. Administration lateral sclerosis (ALS). HD is a neurodegenerative brain disor- of FTY720, a sphingosine analog metabolized in the tissue into der involving striatum and cortex and manifesting itself in mo- a S1P receptor modulator, protected against neurodegeneration tor and cognitive disturbances. It is caused by a dominant mu- and behavioral defects in mouse PD models induced by MPTP, tation, a triplet expansion in the huntingtin (HTT) gene. HD 6-hydroxydopamine, or rotenone—via S1PR1 and probably appears to disturb sphingolipid metabolism; increased SGPL1 Akt [194, 207, 208]. In turn, secreted ASN can inhibit S1PR1 protein has been observed in the cortex and striatum of ad- signaling and disturb the receptor’s localization in lipid rafts vanced HD postmortem brains, accompanied by striatal reduc- [130]. It is worth mentioning, however, that FTY720 failed to tioninSPHK1[226]. Most data, however, has been obtained offer protection in a model of PD induced by subacute (5 days) from animal models. Results suggest, similarly to other neuro- MPTP administration [209]. degenerative disorders, that imbalance in sphingolipid concen- The PD-linked changes may exert influence not only on neu- trations and enzyme expression levels occurs in early stages of ronal survival and phenotype, but also potentially on the central the disease development [227–230]. SPTLC1 and CERS1 were mechanisms of PD pathology. Sphingomyelin has been demon- reduced in the brains of R6/2 mice [227], a HD model mice strated to modify the expression levels of ASN [210]. transgenic for the first exon of huntingtin harboring ca. 160 Degradation of overexpressed or otherwise pathologically altered CAG repeats [231]. The altered sphingolipid metabolism has ASN may be dependent on the sphingomyelinase [211]. also been noted in a variety of other cellular and animal HD Pharmacological inhibition of SPHK1/-2 activities in cells treated models [226], although the changes seem to be less clearly with low concentrations of MPP+ leads to enhanced secretion of weighted towards cell death than in, e.g., AD [228]. However, ASN, which may strengthen the significance of the new, still the reduced S1P levels observed in R6/2 mice [226]appearto underestimated mechanism of Parkinsonian pathology [45]. be a relevant potential therapeutic target, as FTY720 has been Outside the CNS, FTY720 also reduced ASN burden in the demonstrated to improve neuronal activity, reduce brain atro- enteric nervous system, improving gut motility whose reduction phy, improve motor function, and increase R6/2 animal survival is an early peripheral symptom in PD [212]. Interestingly, [232]. Moreover, S1PR agonists increased huntingtin phos- pramipexole (a dopamine D2/D3 receptor agonist) reversed phorylation and reduced its aggregation [232, 233]. FTY720 SPHK1 inhibition in the MPTP model [194], suggesting further also increased the levels of brain-derived neurotrophic factor interactions between sphingolipid and dopamine signaling. (BDNF) levels and mitigated the upregulation of NF-κB, Glucocerebrosidase (GBA) is a lysosomal enzyme that pro- iNOS, and TNF-α that would otherwise lead to the potentially duces ceramide from glucocerebroside (glucosylceramide) [196]. neurotoxic activation of astrocytes; FTY720 thus preserved GBA deficiency/mutations are among top genetic contributors to synaptic plasticity and memory in R6/1 mice (another model the development of PD [196, 213, 214] and are statistically as- with lower number of glutamine repeats in the first huntingtin sociated with Parkinson’s disease [215, 216], contributing to its exon) [234]. The S1PR5 stimulator A-971432 preserved blood- early development, rapid progression, and presence of additional brain barrier integrity in R6/2 mice [233]. These results have led psychiatric symptoms [217, 218]. Interestingly, β- to proposal of S1P-modulating therapy of HD [235]. glucocerebrosidase activity is reduced in the cerebrospinal fluid ALS is a neurodegenerative disorder encompassing motor (CSF) of PD patients even if they do not carry any GBA1 muta- neuron degeneration, muscle wasting, and paralysis, charac- tions [219]. Variants in the GBA gene may be highly useful in the terized by severe deregulation of metabolism, including lipid prediction of PD course [220]. Accumulation of glucosyl com- metabolism [236]. Ceramides and their glucosyl and lactosyl pounds and cholesterol has attracted most attention as the derivatives are increased in ALS patient spinal cords [237]. pathomechanism in GBA mutations/deficiency [221], but chang- SOD mutant mouse model of ALS displays disturbances in es in ceramide levels cannot be excluded as an important con- the expression of genes related to immune regulation, tributing factor [214, 222]. The enzyme is important in ASN exosomal secretion, or lysosomes. Importantly, disturbed degradation [223] and appears to protect against ASN aggrega- levels of ceramides and sphingosine were noted to correlate tion [224]. Additionally, PD patients not carrying the GBA mu- with disease severity along with the expression of SPHK1, or tation also display elevated glucosylceramides in their plasma SGPP2 and sphingolipids—sphingosine and ceramides [198]. Small-molecule GBA chaperones have been suggested (d18:1/26:0) [238]. Increased glucosyl ceramide synthase as a possible means of therapy in synucleinopathies [225]. (GCS) expression might hamper normalization of oxidative metabolism and motor recovery [236]. Also in this case, FTY720 improved neurological scores and survival in SOD Sphingolipids in Huntington’sDisease mutant mice [239]. It modified the mRNA expression of, e.g., and Amyotrophic Lateral Sclerosis iNOS (reduced by the treatment), ARG1 (increased), BDNF (increased), and interleukin genes (IL-1β reduced, IL-10 in- In recent years, data has been accumulating on the engagement creased), despite administration starting in the symptomatic of sphingolipids in Huntington’s disease (HD) and amyotrophic phase9 [23 ]. Mol Neurobiol (2019) 56:5436–5455 5445

MicroRNA Signaling and Bioactive shown to be altered in AD- and PD-affected brain cells and Sphingolipids in Neurodegenerative tissues [177, 244–250]. Examples of target mRNAs include Disorders [244, 245, 248]: miRNAs are increasingly viewed as central regulators of & Immune regulators such as NF-κB, interleukins 4 and 17a, neuronal homoeostasis, and their causal roles in neurode- interleukin-1 receptor-associated kinase-1 (IRAK-1), or generative disorders are rapidly gaining attention. complement factor-H (CFH) Deregulation of miRNA-based gene expression control & Neuronal activity/synaptic plasticity/scaffold genes such may be a novel disease mechanism, but also delivers po- as glutamate receptor genes NR2A, GluR1, synaptobrevin tentially valuable biomarkers of its development 2, and synaptotagmin 1 [240–242]. Considerable research interest has been gener- & Genes coding for glycolysis and oxidative phosphoryla- ated concerning the role of miRNAs in the neuropatholo- tion proteins such as succinate dehydrogenase complex C, gy of bioactive sphingolipids in several progressive age- ubiquinol-cytochrome c reductase binding protein and related human neuropathological diseases, and especially ubiquinol-cytochrome c reductase complex III subunit how specific miRNAs may contribute to the dynamic VII (UQCRB and UQCRQ, respectively), or molecular-genetic processes involving aberrant ceramide/ phosphofructokinase-1 C1P/S1P metabolism in both AD and PD. In humans, & Amyloidogenesis-linked genes such as the membrane protein miRNAsareafamilyof18–22 nt single-stranded RNAs tetraspanin 12 (TSPAN12), or the master postsynaptic that posttranslationally interact with, and regulate, the ex- membrane-organizing ankyrin-cytoskeletal protein SHANK3 pression of mature mRNAs. Single upregulated miRNA can target multiple mRNAs to reduce their expression, Put another way, specific pathology-linked miRNAs ap- and multiple miRNAs can target a single mRNA pear to regulate a large number of plasma membrane- [243–245]. Whenever progressive neurodegeneration is resident and plasma membrane-organizing components encountered in central nervous tissues undergoing patho- whose character is defined by sphingolipid composition, turn- logical change, progressive neuronal atrophy and brain over, and metabolism. cell death the NF-κB-sensitive, pro-inflammatory and po- Many CNS-abundant miRNAs have, in addition, important tentially pathogenic miRNA species such as miRNA-34a, regulatory functions in the expression of enzymes involved in miRNA-146a, miRNA-155, and several others have been the generation of ceramide, sphingosine, C1P, S1P, and/or their showntobeabundant(andreadilydetectablebyhybrid- receptors, both in healthy brain aging and in neurological dis- ization methodologies) in the cytoplasm of degenerating ease. For example, the pro-inflammatory and rapidly induced neurons, as well as in both the extracellular fluid (ECF) NF-κB-regulated miRNA-155 (encoded in humans at chr and CSF, which is contiguous with ECF. While these 21q21.3) has been shown to regulate biosynthesis of the miRNAs are normally required for the homeostatic oper- S1PR1 which functions in the amelioration of pathogenic in- ation of brain cellular and membrane-signaling functions, flammation in systemic autoimmune disease [246–248, 251]. their upregulation and persistence in deteriorating nervous Interestingly, a five-member cluster of miRNAs encoded on tissues and the nature of their interaction with biological human 21 that includes let-7c, miRNA-99a, membranes is associated with, and indicative of, the prop- miRNA-125b, miRNA-155, and miRNA-802 may help ex- agation and spreading of neurodegenerative disease. plain the complex phenotypic diversity of trisomy 21 These miRNAs may be a diagnostic tool for the cytoplas- (Down’s syndrome; DS) and the strong linkage between DS mic status of brain cells at risk for neurodegeneration and the aberrant sphingolipid and ceramide metabolism associ- [241–243]. ated with trisomy 21 (DS) neuropathology [252–254]. In turn, the upregulated microRNAs such as miRNA-34a, Interestingly, some of the most recent brain biolipid research miRNA-146a, and miRNA-155 appear to antagonize both describes the association between neurotoxins secreted by the individual mRNAs and small families of functionally related human gastrointestinal (GI) tract microbiome and inflammatory mRNAs and, in doing so, affect entire systems of CNS- neurodegeneration of nervous tissues, a complex pathogenic membrane-relevant genes and plasma membrane processes. process that is certain to involve CNS sphingolipid composi- Indeed, overexpression of miRNA-34a, miRNA-146a, and/ tion, their organization, and interactive metabolism [255, 256]. or miRNA-155 have been shown to affect the expression of As mentioned earlier, sphingolipids are key regulators of a large number of genes normally involved in glucose metab- exosomal secretion. Their roles include exosome formation, en- olism, innate-immune regulation, membrane integrity, normal capsulation, and miRNA shuttling across the plasma membrane vascular function and endothelial cell permeability, oxidative [257]. Exosomes and other extracellular vesicles secreted into the phosphorylation, synaptic plasticity, and exosome generation, extracellular space from both neuronal and glial cells are enriched encapsulation, and release. All of these processes have been with the sphingolipid ceramide, as well as other more complex 5446 Mol Neurobiol (2019) 56:5436–5455 glycosphingolipids such as gangliosides, and may also be changes in astrocyte phenotype such as reduced expression of enriched in various species of pathogenic or “communicating” excitatory amino acid transporter 2 or peroxisome proliferator- miRNAs [177, 243–245]. Such exosomal vesicle-bound activated receptor gamma coactivator 1α andtoastrogliosis miRNAs: (a) should be reflective of the sphingolipid and which likely contributes to the neuron loss [269]. miRNA composition of the brain cell cytoplasm from which they were originally derived; (b) may serve the role as a novel form of intercellular communication among brain cells; (c) may carry Concluding Remarks selective miRNA ‘cargos’ that regulate both bioactive sphingosine/S1P and ceramide/C1P metabolism as well as other Neurodegenerative disorders belong to the most widespread, miRNA-mRNA targets in adjacent cells; (d) have been implicat- devastating, and uncontrollable diseases. Alzheimer’s, ed in the inter-neuronal “spreading” of pathogenic signals via Parkinson’s, and Huntington’s diseases and amyotrophic lateral “paracrine” and related secretory effects in the diseased and sclerosis are, like physiological aging, increasingly associated neuro-degenerating brain; (e) have considerable potential for be- with pronounced disturbances in the metabolism of bioactive ing clinically useful as a predictor and non-invasive diagnostic sphingolipids (generally tending to augment the pro-apoptotic marker for AD and/or PD; and (f) may provide a “molecular- ceramide signaling at the expense of survival signals mediated genetic” signature for a defined group of miRNAs associated by S1P). However, recent findings suggest a more complex pic- with a particular neurological disease [177, 243–248]. ture, creating the need for refinement of current knowledge on of Recently, data on the engagement of miRNA-based gene the roles of S1P and ceramides in the various cell death modes. regulation in HD and ALS begun to accumulate. Postmortem The early appearance of sphingolipid alterations suggests their HD cortex samples from Brodmann’s area 4 reveal disturbed engagement in upstream steps of disease development. These miRNA expression (reduced miR-9, miR-9*, miR-29b, miR- observations raise hopes for identification of therapeutic targets 124a, miR-132) that might stem from loss of huntingtin- that would allow reaching beyond the current symptomatic treat- transcription factor interaction in neuronal cells [258]. ments. They also should help in the identification of highly us- Numerous deregulated circulating miRNAs have been found able biomarkers for the still elusive goal of early diagnosis. in HD cases and might reflect not only the ongoing neurode- Besides cell survival/death signaling, the roles of generation but also altered communication with the periphery sphingolipids are more obscure. Their significance in the metab- [259]. Links between altered miRNAs and perturbations in ap- olism of AβPP/Aβ andASN,withbothproteins’ physiological optotic and signaling have been proposed as a possi- roles still unclear, needs extensive insights before conclusions ble mechanism of cell loss in HD [260]. The R6/2 mouse HD can be drawn. Similar is the significance of sphingolipids’ links model displays reduction in miRNA-34a-5p, a member of with secretion mechanisms which can affect the spread of aggre- miRNA-34 family that is engaged in p53- and SIRT1- gating proteins, death/survival signals, or metabolic regulators dependent modulation of cell cycle, senescence, and apoptosis such as noncoding RNAs. The significance of miRNAs for neu- [261]. A series of mouse models with various numbers of CAG rodegenerative disorders, although gaining increasing recogni- repeats in the huntingtin gene has shown a repeat number- and tion in the field, is still largely uncharacterized. brain part-dependent alteration in miRNA transcriptome (in- A final question is that about the availability of therapeutic cluding miRNAs engaged in neuronal development/survival tools to manipulate the extremely complex network of [262]). Altered miRNA levels in blood plasma and CSF have sphingolipid metabolism. Some of the most basic needs may be been proposed as HD biomarkers [263, 264]. met with currently available repurposed drugs such as fingolimod; The engagement of microRNAs in the pathology not only of however, it is highly possible that exploitation of sphingolipids as neurons but also muscles is relatively better characterized in ALS therapeutic targets (as opposed to their use in diagnosis) may [265]. A high-throughput next-generation sequencing project has require significant expansion of the current toolset. identified reduction in the blood levels of 38 miRNAs in sporadic ALS patients, including let-7 and miR-26 families. The pattern of Funding The work was supported by the National Science Centre (PL) – reductions was dependent on the disease phenotypic expression Grant No. 2013/11/N/NZ4/02233 (Chapters 1 4), an unrestricted grant to the LSU Eye Center from Research to Prevent Blindness (RPB) and a and progression rate, making them potentially useful for diagnos- National Institute of Health (NIH) grant NIA AG038834 and Mossakowski tic purposes [266]. In turn, upregulation of miR-223-3p, miR- Medical Research Centre Polish Academy of Sciences –T. No. 7 (Chapter 5). 326, and miR-338-3p observed in tissue bank neuromuscular junction samples of ALS patients may disturb HIF-1 and brain- Compliance with Ethical Standards derived neurotrophic factor signaling [267]. Disruption of the intraneuronal localization of RNAi machinery (leading to Conflict of Interest The authors declare that they have no conflict of deregulated axonal protein synthesis) has also been noted as an interest. important aspect of ALS pathology [268]. Moreover, Open Access This article is distributed under the terms of the Creative degenerating/dying neurons release miRNA-218 which leads to Commons Attribution 4.0 International License (http:// Mol Neurobiol (2019) 56:5436–5455 5447 creativecommons.org/licenses/by/4.0/), which permits unrestricted use, 15. Spassieva SD, Ji X, Liu Y et al (2016) Ectopic expression of distribution, and reproduction in any medium, provided you give appro- in neurons suppresses neurodegeneration in- priate credit to the original author(s) and the source, provide a link to the duced by deficiency. Proc Natl Acad Sci U S Creative Commons license, and indicate if changes were made. A 113:5928–5933. https://doi.org/10.1073/pnas.1522071113 16. Zhao L, Spassieva SD, Jucius TJ et al (2011) A deficiency of ceramide biosynthesis causes cerebellar purkinje cell neurodegen- eration and lipofuscin accumulation. PLoS Genet 7:e1002063. Publisher’sNoteSpringer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations. https://doi.org/10.1371/journal.pgen.1002063 17. Eberle M, Ebel P, Mayer CA et al (2015) Exacerbation of experimen- tal autoimmune encephalomyelitis in ceramide synthase 6 knockout References mice is associated with enhanced activation/migration of neutrophils. ImmunolCellBiol93:825–836. https://doi.org/10.1038/icb.2015.47 18. McGovern MM, Avetisyan R, Sanson B-J, Lidove O (2017) 1. Winblad B, Amouyel P, Andrieu S et al (2016) Defeating Disease manifestations and burden of illness in patients with acid ’ Alzheimer s disease and other dementias: a priority for sphingomyelinase deficiency (ASMD). Orphanet J Rare Dis 12: – European science and society. Lancet Neurol 15:455 532. 41. https://doi.org/10.1186/s13023-017-0572-x https://doi.org/10.1016/S1474-4422(16)00062-4 19. Sikora J, Dworski S, Jones EE et al (2017) Acid ceramidase defi- 2. Brunkhorst R, Vutukuri R, Pfeilschifter W (2014) Fingolimod for ciency in mice results in a broad range of central nervous system the treatment of neurological diseases-state of play and future abnormalities. Am J Pathol 187:864–883. https://doi.org/10.1016/ perspectives. Front Cell Neurosci 8:283. https://doi.org/10.3389/ j.ajpath.2016.12.005 fncel.2014.00283 20. Wang K, Xu R, Schrandt J et al (2015) Alkaline ceramidase 3 defi- 3. Gao Y-L, Wang N, Sun F-R et al (2018) Tau in neurodegenerative ciency results in Purkinje cell degeneration and cerebellar Ataxia due disease. Ann Transl Med 6:175. https://doi.org/10.21037/atm.2018.04. to dyshomeostasis of sphingolipids in the brain. PLoS Genet 11: 23 e1005591. https://doi.org/10.1371/journal.pgen.1005591 4. Jęśko H, Lenkiewicz AM, Adamczyk A (2017) Treatments and 21. Lu M-H, Takemoto M, Watanabe K et al (2012) Deficiency of compositions targeting α-synuclein: a patent review (2010-2016). sphingomyelin synthase-1 but not sphingomyelin synthase-2 Expert Opin Ther Pat 27:427–438. https://doi.org/10.1080/ causes hearing impairments in mice. J Physiol 590:4029–4044. 13543776.2017.1261112 https://doi.org/10.1113/jphysiol.2012.235846 5. de Lau LM, Breteler MM (2006) Epidemiology of Parkinson’s 22. Wang M, Uchiumi O, Ogiso H et al (2017) Stressful learning disease. Lancet Neurol 5:525–535. https://doi.org/10.1016/ paradigm precludes manifestation of cognitive ability in S1474-4422(06)70471-9 sphingomyelin synthase-2 knockout mice. Behav Brain Res 319: 6. Aarsland D, Creese B, Politis M et al (2017) Cognitive decline in 25–30. https://doi.org/10.1016/j.bbr.2016.11.010 Parkinson disease. Nat Rev Neurol 13:217–231. https://doi.org/ 23. Kono M, Belyantseva IA, Skoura A et al (2007) Deafness and stria 10.1038/nrneurol.2017.27 vascularis defects in S1P2 receptor-null mice. J Biol Chem 282: – 7. Nakamura S, Fukai E, Miya S et al (2011) Sphingolipid signaling 10690 10696. https://doi.org/10.1074/jbc.M700370200 and neuronal function. Chem Phys Lipids 164:S9. https://doi.org/ 24. MacLennan AJ, Carney PR, Zhu WJ et al (2001) An essential role 10.1016/J.CHEMPHYSLIP.2011.05.023 for the H218/AGR16/Edg-5/LP(B2) sphingosine 1-phosphate re- ceptor in neuronal excitability. Eur J Neurosci 14:203–209 8. Jęśko H, Stępień A, Lukiw WJ, Strosznajder RP (2018) The cross- talk between sphingolipids and insulin-like growth factor signal- 25. Cutler RG, Thompson KW, Camandola S et al (2014) ing: significance for aging and neurodegeneration. Mol Neurobiol Sphingolipid metabolism regulates development and lifespan in Caenorhabditis elegans. Mech Ageing Dev 143–144:9–18. https://doi.org/10.1007/s12035-018-1286-3 https://doi.org/10.1016/j.mad.2014.11.002 9. Mencarelli C, Martinez-Martinez P (2013) Ceramide function in 26. Yang Q, Gong Z-J, Zhou Y et al (2010) Role of Drosophila alka- the brain: when a slight tilt is enough. Cell Mol Life Sci 70:181– line ceramidase (Dacer) in Drosophila development and longevity. 203. https://doi.org/10.1007/s00018-012-1038-x Cell Mol Life Sci 67:1477–1490. https://doi.org/10.1007/s00018- 10. Maceyka M, Spiegel S (2014) Sphingolipid metabolites in inflam- 010-0260-7 – matory disease. Nature 510:58 67. https://doi.org/10.1038/ 27. Jové M, Naudí A, Gambini J et al (2017) A stress-resistant nature13475 Lipidomic signature confers extreme longevity to humans. J 11. Airola MV,Hannun YA (2013) Sphingolipid metabolism and neu- Gerontol A Biol Sci Med Sci 72:30–37. https://doi.org/10.1093/ tral sphingomyelinases. In: Gulbins E., Petrache I. (eds) gerona/glw048 Sphingolipids: Basic Science and Drug Development. Handb 28. Couttas TA, Kain N, Suchowerska AK et al (2016) Loss of cer- – Exp Pharmacol 215:57 76. Springer, Vienna. https://doi.org/10. amide synthase 2 activity, necessary for biosynthesis, pre- 1007/978-3-7091-1368-4_3 cedes tau pathology in the cortical pathogenesis of Alzheimer’s 12. Motyl J, Strosznajder JB (2018) Sphingosine kinase 1/sphingo- disease. Neurobiol Aging 43:89–100. https://doi.org/10.1016/j. sine-1-phosphate receptors dependent signalling in neurodegener- neurobiolaging.2016.03.027 ative diseases. The promising target for neuroprotection in 29. Han XM, Holtzman D, McKeel DW et al (2002) Substantial sulfatide Parkinson’s disease. Pharmacol Rep 70:1010–1014. https://doi. deficiency and ceramide elevation in very early Alzheimer’s disease: org/10.1016/j.pharep.2018.05.002 potential role in disease pathogenesis. J Neurochem 82:809–818 13. Hagen N, Hans M, Hartmann D et al (2011) Sphingosine-1- 30. Giussani P, Tringali C, Riboni L et al (2014) Sphingolipids: key phosphate links glycosphingolipid metabolism to neurodegenera- regulators of apoptosis and pivotal players in cancer drug resis- tion via a calpain-mediated mechanism. Cell Death Differ 18: tance. Int J Mol Sci 15:4356–4392. https://doi.org/10.3390/ 1356–1365. https://doi.org/10.1038/cdd.2011.7 ijms15034356 14. Rotthier A, Auer-Grumbach M, Janssens K et al (2010) Mutations 31. O’Sullivan S, Dev KK (2017) Sphingosine-1-phosphate receptor in the SPTLC2 subunit of serine palmitoyltransferase cause hered- therapies: advances in clinical trials for CNS-related diseases. itary sensory and autonomic neuropathy type I. Am J Hum Genet Neuropharmacology 113:597–607. https://doi.org/10.1016/j. 87:513–522. https://doi.org/10.1016/j.ajhg.2010.09.010 neuropharm.2016.11.006 5448 Mol Neurobiol (2019) 56:5436–5455

32. Granado MH, Gangoiti P, Ouro A et al (2009) Ceramide 1- glomerular mesangial cells exposed to high glucose. Cell Signal phosphate (C1P) promotes cell migration involvement of a spe- 26:629–638. https://doi.org/10.1016/j.cellsig.2013.12.002 cific C1P receptor. Cell Signal 21:405–412. https://doi.org/10. 49. Wegner M-S, Wanger RA, Oertel S et al (2014) Ceramide 1016/j.cellsig.2008.11.003 synthases CerS4 and CerS5 are upregulated by 17β-estradiol 33. Abdelbaset-Ismail A, Cymer M, Borkowska-Rzeszotek S, et al (2018) and GPER1 via AP-1 in human breast cancer cells. Biochem Bioactive phospholipids enhance migration and adhesion of human Pharmacol 92:577–589. https://doi.org/10.1016/j.bcp.2014.10. leukemic cells by inhibiting Heme oxygenase 1 (HO-1) and inducible 007 nitric oxygenase synthase (iNOS) in a p38 MAPK-dependent manner. 50. O’Neill SM, Houck KL, Yun JK et al (2011) AP-1 binding tran- Stem Cell Rev https://doi.org/10.1007/s12015-018-9853-6 scriptionally regulates human neutral ceramidase. Arch Biochem 34. Nagahashi M, Takabe K, Terracina KP et al (2014) Sphingosine-1- Biophys 511:31–39. https://doi.org/10.1016/j.abb.2011.04.009 phosphate transporters as targets for cancer therapy. Biomed Res 51. Manna SK, Sah NK, Aggarwal BB (2000) Protein tyrosine kinase Int 2014:1–7. https://doi.org/10.1155/2014/651727 p56lck is required for ceramide-induced but not tumor necrosis 35. Farooqui AA, Ong W-Y,Farooqui T (2010) Lipid mediators in the factor-induced activation of NF-kappa B, AP-1, JNK, and apopto- nucleus: their potential contribution to Alzheimer’sdisease. sis. J Biol Chem 275:13297–13306 – Biochim Biophys Acta 1801:906 916. https://doi.org/10.1016/j. 52. Dong Y-F, Guo R-B, Ji J et al (2018) S1PR3 is essential for phos- bbalip.2010.02.002 phorylated fingolimod to protect astrocytes against oxygen- 36. Rutherford C, Childs S, Ohotski J et al (2013) Regulation of cell glucose deprivation-induced neuroinflammation via inhibiting survival by sphingosine-1-phosphate receptor S1P1 via reciprocal TLR2/4-NFκB signalling. J Cell Mol Med 22:3159–3166. ERK-dependent suppression of Bim and PI-3-kinase/protein ki- https://doi.org/10.1111/jcmm.13596 nase C-mediated upregulation of Mcl-1. Cell Death Dis 4:e927. 53. Klotz L-O, Sánchez-Ramos C, Prieto-Arroyo I et al (2015) Redox https://doi.org/10.1038/cddis.2013.455 regulation of FOXO transcription factors. Redox Biol 6:51–72. šć Š ć 37. Jazvin ak Jembrek M, Hof PR, imi G (2015) Ceramides in https://doi.org/10.1016/j.redox.2015.06.019 ’ Alzheimer s disease: key mediators of neuronal apoptosis induced 54. Maiese K (2018) Forkhead transcription factors: formulating a β by oxidative stress and A accumulation. Oxidative Med Cell FOXO target for cognitive loss. Curr Neurovasc Res 14:415– Longev 2015:346783. https://doi.org/10.1155/2015/346783 420. https://doi.org/10.2174/1567202614666171116102911 38. Snook CF, Jones JA, Hannun YA (2006) Sphingolipid-binding 55. Kogot-Levin A, Saada A (2014) Ceramide and the mitochondrial proteins. Biochim Biophys Acta 1761:927–946. https://doi.org/ respiratory chain. Biochimie 100:88–94. https://doi.org/10.1016/J. 10.1016/j.bbalip.2006.06.004 BIOCHI.2013.07.027 39. Xu R, Wang K, Mileva I et al (2016) Alkaline ceramidase 2 and its 56. Stoica BA, Movsesyan VA, Knoblach SM, Faden AI (2005) bioactive product sphingosine are novel regulators of the DNA Ceramide induces neuronal apoptosis through mitogen-activated damage response. Oncotarget 7:18440–18457. https://doi.org/10. protein kinases and causes release of multiple mitochondrial pro- 18632/oncotarget.7825 teins. Mol Cell Neurosci 29:355–371. https://doi.org/10.1016/j. 40. Gómez-Muñoz A, Kong JY, Salh B, Steinbrecher UP (2004) mcn.2005.02.009 Ceramide-1-phosphate blocks apoptosis through inhibition of acid 57. Sentelle RD, Senkal CE, Jiang W et al (2012) Ceramide targets sphingomyelinase in macrophages. J Lipid Res 45:99–105. autophagosomes to mitochondria and induces lethal mitophagy. https://doi.org/10.1194/jlr.M300158-JLR200 Nat Chem Biol 8:831–838. https://doi.org/10.1038/nchembio. 41. Granado MH, Gangoiti P, Ouro A et al (2009) Ceramide 1- 1059 phosphate inhibits serine palmitoyltransferase and blocks apopto- sis in alveolar macrophages. Biochim Biophys Acta 1791:263– 58. Jiang W, Ogretmen B (2014) Autophagy paradox and ceramide. – 272. https://doi.org/10.1016/j.bbalip.2009.01.023 Biochim Biophys Acta - Mol Cell Biol Lipids 1841:783 792. 42. Rivera I-G, Ordoñez M, Presa N et al (2015) Sphingomyelinase https://doi.org/10.1016/j.bbalip.2013.09.005 D/ceramide 1-phosphate in cell survival and inflammation. Toxins 59. Hait NC, Allegood J, Maceyka M et al (2009) Regulation of his- (Basel) 7:1457–1466. https://doi.org/10.3390/toxins7051457 tone acetylation in the nucleus by sphingosine-1-phosphate. – 43. Choi JW, Chun J (2013) Lysophospholipids and their receptors in Science 325:1254 1257. https://doi.org/10.1126/science.1176709 the central nervous system. Biochim Biophys Acta - Mol Cell Biol 60. Schmitt HM, Schlamp CL, Nickells RW (2018) Targeting Lipids 1831:20–32. https://doi.org/10.1016/j.bbalip.2012.07.015 HDAC3 activity with RGFP966 protects against retinal ganglion 44. Couttas TA, Kain N, Tran C et al (2018) Age-dependent changes cell nuclear atrophy and apoptosis after optic nerve injury. J Ocul – to sphingolipid balance in the human hippocampus are gender- Pharmacol Ther 34:260 273. https://doi.org/10.1089/jop.2017. specific and may sensitize to neurodegeneration. J Alzheimers 0059 Dis 63:503–514. https://doi.org/10.3233/JAD-171054 61. Jeong M-H, Ko H, Jeon H et al (2016) Delphinidin induces apo- 45. Pyszko JA, Strosznajder JB (2014) The key role of sphingosine ptosis via cleaved HDAC3-mediated p53 acetylation and oligo- kinases in the molecular mechanism of neuronal cell survival and merization in prostate cancer cells. Oncotarget 7:56767–56780. death in an experimental model of Parkinson’s disease. Folia https://doi.org/10.18632/oncotarget.10790 Neuropathol 52:260–269 62. Lee S-H, Seo J, Park S-Y et al (2018) Programmed cell death 5 46. Motyl J, Przykaza Ł, Kosson P, Boguszewski P, Strosznajder J suppresses AKT-mediated cytoprotection of endothelium. Proc (2015) P.5.c.002 sphingosine kinase 1 mediated signalling in Natl Acad Sci 115:201712918. https://doi.org/10.1073/pnas. Parkinson’s disease animal model. Neuroprotective effect of 1712918115 fingolimod and a D2/D3 dopamine receptor agonist. Eur 63. Malaplate-Armand C, Florent-Béchard S, Youssef I et al (2006) Neuropsychopharmacol 25:S586–S587. https://doi.org/10.1016/ Soluble oligomers of amyloid-beta peptide induce neuronal apo- S0924-977X(15)30824-5 ptosis by activating a cPLA2-dependent sphingomyelinase-cer- 47. Czubowicz K, Strosznajder R (2014) Ceramide in the molecular amide pathway. Neurobiol Dis 23:178–189. https://doi.org/10. mechanisms of neuronal cell death. The role of sphingosine-1- 1016/j.nbd.2006.02.010 phosphate. Mol Neurobiol 50:26–37. https://doi.org/10.1007/ 64. Pchejetski D, Kunduzova O, Dayon A et al (2007) Oxidative s12035-013-8606-4 stress-dependent sphingosine kinase-1 inhibition mediates mono- 48. Huang K, Huang J, Chen C et al (2014) AP-1 regulates sphingo- amine oxidase A-associated cardiac cell apoptosis. Circ Res 100: sine kinase 1 expression in a positive feedback manner in 41–49. https://doi.org/10.1161/01.RES.0000253900.66640.34 Mol Neurobiol (2019) 56:5436–5455 5449

65. Couttas TA, Kain N, Daniels B et al (2014) Loss of the neuropro- 81. Subramaniam S, Unsicker K (2010) ERK and cell death: ERK1/2 tective factor sphingosine 1-phosphate early in Alzheimer’sdis- in neuronal death. FEBS J 277:22–29. https://doi.org/10.1111/j. ease pathogenesis. Acta Neuropathol Commun 2:9. https://doi. 1742-4658.2009.07367.x org/10.1186/2051-5960-2-9 82. Wang L, Cheng B-F, Yang H-J et al (2015) NF-κBprotectshuman 66. Czubowicz K, Cieślik M, Pyszko J et al (2015) Sphingosine-1- neuroblastoma cells from nitric oxide-induced apoptosis through phosphate and its effect on glucose deprivation/glucose reload upregulating biglycan. Am J Transl Res 7:1541–1552 stress: from gene expression to neuronal survival. Mol Neurobiol 83. Kim A, Nam YJ, Lee CS (2017) Taxifolin reduces the cholesterol 51:1300–1308. https://doi.org/10.1007/s12035-014-8807-5 oxidation product-induced neuronal apoptosis by suppressing the 67. Van Brocklyn JR, Williams JB (2012) The control of the balance Akt and NF-κB activation-mediated cell death. Brain Res Bull between ceramide and sphingosine-1-phosphate by sphingosine 134:63–71. https://doi.org/10.1016/j.brainresbull.2017.07.008 kinase: oxidative stress and the seesaw of cell survival and death. 84. Schwartz NU, Linzer RW, Truman J-P et al (2018) Decreased Comp Biochem Physiol B Biochem Mol Biol 163:26–36. https:// ceramide underlies mitochondrial dysfunction in Charcot-Marie- doi.org/10.1016/j.cbpb.2012.05.006 Tooth 2F. FASEB J 32:1716– 1728. https://doi.org/10.1096/fj. 68. Chen W, Lu H, Yang J et al (2016) Sphingosine 1-phosphate in 201701067R metabolic syndrome (review). Int J Mol Med 38:1030–1038. 85. van Blitterswijk WJ, van der Luit AH, Veldman RJ et al (2003) https://doi.org/10.3892/ijmm.2016.2731 Ceramide: second messenger or modulator of membrane structure 69. Hasegawa Y, Suzuki H, Sozen T et al (2010) Activation of sphin- and dynamics? Biochem J 369:199–211. https://doi.org/10.1042/ gosine 1-phosphate receptor-1 by FTY720 is neuroprotective after BJ20021528 ischemic stroke in rats. Stroke 41:368–374. https://doi.org/10. 86. Mesicek J, Lee H, Feldman Tet al (2010) Ceramide synthases 2, 5, 1161/STROKEAHA.109.568899 and 6 confer distinct roles in radiation-induced apoptosis in HeLa 70. Safarian F, Khallaghi B, Ahmadiani A, Dargahi L (2015) cells. Cell Signal 22:1300–1307. https://doi.org/10.1016/j.cellsig. Activation of S1P1 receptor regulates PI3K/Akt/FOXO3a pathway 2010.04.006 in response to oxidative stress in PC12 cells. J Mol Neurosci 56: 87. Cruickshanks N, Roberts JL, Bareford MD et al (2015) 177–187. https://doi.org/10.1007/s12031-014-0478-1 Differential regulation of autophagy and cell viability by ceramide 71. Moloney AM, Griffin RJ, Timmons S et al (2010) Defects in IGF- species. Cancer Biol Ther 16:733–742. https://doi.org/10.1080/ 1 receptor, insulin receptor and IRS-1/2 in Alzheimer’s disease 15384047.2015.1026509 indicate possible resistance to IGF-1 and insulin signalling. 88. Ghosh S, Bhattacharyya S, Sirkar M et al (2002) Leishmania Neurobiol Aging 31:224–243. https://doi.org/10.1016/j. donovani suppresses activated protein 1 and NF-kappaB activa- neurobiolaging.2008.04.002 tion in host macrophages via ceramide generation: involvement of 72. Lauzon M-A, Daviau A, Marcos B, Faucheux N (2015) Growth extracellular signal-regulated kinase. Infect Immun 70:6828–6838 factor treatment to overcome Alzheimer’s dysfunctional signaling. 89. Colombini M (2017) Ceramide channels and mitochondrial outer Cell Signal 27:1025–1038. https://doi.org/10.1016/j.cellsig.2015. membrane permeability. J Bioenerg Biomembr 49:57–64. https:// 02.018 doi.org/10.1007/s10863-016-9646-z 73. Hsu C-K, Lee I-T, Lin C-C et al (2015) Sphingosine-1-phosphate 90. Liu Z, Li T, Li P et al (2015) The ambiguous relationship of mediates COX-2 expression and PGE2 /IL-6 secretion via c-Src- oxidative stress, tau hyperphosphorylation, and autophagy dys- dependent AP-1 activation. J Cell Physiol 230:702–715. https:// function in Alzheimer’s disease. Oxidative Med Cell Longev doi.org/10.1002/jcp.24795 2015:352723. https://doi.org/10.1155/2015/352723 74. Alvarez SE, Harikumar KB, Hait NC et al (2010) Sphingosine-1- 91. Zhang L, Wang L, Wang R et al (2017) Evaluating the effective- phosphate is a missing cofactor for the E3 ubiquitin ligase TRAF2. ness of GTM-1, rapamycin, and carbamazepine on autophagy and Nature 465:1084–1088. https://doi.org/10.1038/nature09128 Alzheimer disease. Med Sci Monit 23:801–808 75. Xia P, Wang L, Moretti PAB et al (2002) Sphingosine kinase 92. Šmidák R, Köfeler HC, Hoeger H, Lubec G (2017) interacts with TRAF2 and dissects tumor necrosis factor-alpha Comprehensive identification of age-related lipidome changes signaling. J Biol Chem 277:7996–8003. https://doi.org/10.1074/ in rat amygdala during normal aging. PLoS One 12:e0180675. jbc.M111423200 https://doi.org/10.1371/journal.pone.0180675 76. Dai Y, Rahmani M, Dent P, Grant S (2005) Blockade of histone 93. Babenko NA, Garkavenko VV, Storozhenko GV, Timofiychuk deacetylase inhibitor-induced RelA/p65 acetylation and NF- OA (2016) Role of acid sphingomyelinase in the age-dependent kappaB activation potentiates apoptosis in leukemia cells through dysregulation of sphingolipids turnover in the tissues of rats. Gen a process mediated by oxidative damage, XIAP downregulation, Physiol Biophys 35:195–205. https://doi.org/10.4149/gpb_ and c-Jun N-terminal kinase 1 activation. Mol Cell Biol 25:5429– 2015046 5444. https://doi.org/10.1128/MCB.25.13.5429-5444.2005 94. Mielke MM, Bandaru VVR, Han D et al (2015) Factors affecting 77. Wang Y,Yang R, Gu J et al (2015) Cross talk between PI3K-AKT- longitudinal trajectories of plasma sphingomyelins: the Baltimore GSK-3β and PP2A pathways determines tau Longitudinal Study of Aging. Aging Cell 14:112–121. https://doi. hyperphosphorylation. Neurobiol Aging 36:188–200. https://doi. org/10.1111/acel.12275 org/10.1016/j.neurobiolaging.2014.07.035 95. Obeid LM, Hannun YA (2003) Ceramide, stress, and a “lag” in 78. Ghasemi R, Dargahi L, Ahmadiani A (2016) Integrated aging. Sci Aging Knowl Environ 2003:PE27. https://doi.org/10. sphingosine-1 phosphate signaling in the central nervous system: 1126/sageke.2003.39.pe27 from physiological equilibrium to pathological damage. 96. Huang X, Withers BR, Dickson RC (2014) Sphingolipids and Pharmacol Res 104:156–164. https://doi.org/10.1016/j.phrs. lifespan regulation. Biochim Biophys Acta - Mol Cell Biol Lipids 2015.11.006 1841:657–664. https://doi.org/10.1016/j.bbalip.2013.08.006 79. Vesely PW, Staber PB, Hoefler G, Kenner L (2009) Translational 97. Jiang JC, Kirchman PA, Allen M, Jazwinski SM (2004) Suppressor regulation mechanisms of AP-1 proteins. Mutat Res Mutat Res analysis points to the subtle role of the LAG1 ceramide synthase 682:7–12. https://doi.org/10.1016/j.mrrev.2009.01.001 gene in determining yeast longevity. Exp Gerontol 39:999–1009. 80. Sawe N, Steinberg G, Zhao H (2008) Dual roles of the https://doi.org/10.1016/j.exger.2004.03.026 MAPK/ERK1/2 cell signaling pathway after stroke. J Neurosci 98. Kim Y, Sun H (2012) ASM-3 acid sphingomyelinase functions as Res 86:1659–1669. https://doi.org/10.1002/jnr.21604 a positive regulator of the DAF-2/AGE-1 signaling pathway and 5450 Mol Neurobiol (2019) 56:5436–5455

serves as a novel anti-aging target. PLoS One 7:e45890. https:// signaling pathways involved in the control of cell survival. J doi.org/10.1371/journal.pone.0045890 Biol Chem 284:5088–5096. https://doi.org/10.1074/jbc. 99. Osawa Y, Banno Y, Nagaki M et al (2001) TNF-alpha-induced M808934200 sphingosine 1-phosphate inhibits apoptosis through a phos- 116. Banno Y, Takuwa Y, Akao Y et al (2001) Involvement of phos- phatidylinositol 3-kinase/Akt pathway in human hepatocytes. J pholipase D in sphingosine 1-phosphate-induced activation of Immunol 167:173–180 phosphatidylinositol 3-kinase and Akt in Chinese hamster ovary 100. Schubert KM, Scheid MP, Duronio V (2000) Ceramide inhibits cells overexpressing EDG3. J Biol Chem 276:35622–35628. protein kinase B/Akt by promoting dephosphorylation of serine https://doi.org/10.1074/jbc.M105673200 473. J Biol Chem 275:13330–13335 117. Davaille J, Li L, Mallat A, Lotersztajn S (2002) Sphingosine 1- 101. Brown-Borg HM (2015) The somatotropic axis and longevity in phosphate triggers both apoptotic and survival signals for human mice. Am J Physiol Endocrinol Metab 309:E503–E510. https:// hepatic myofibroblasts. J Biol Chem 277:37323–37330. https:// doi.org/10.1152/ajpendo.00262.2015 doi.org/10.1074/jbc.M202798200 102. Selman C, Partridge L, Withers DJ (2011) Replication of extended 118. Cui H, Okamoto Y, Yoshioka K et al (2013) Sphingosine-1-phosphate lifespan phenotype in mice with deletion of insulin receptor sub- receptor 2 protects against anaphylactic shock through suppression of strate 1. PLoS One 6:e16144. https://doi.org/10.1371/journal. endothelial nitric oxide synthase in mice. J Allergy Clin Immunol 132: pone.0016144 1205–1214.e9. https://doi.org/10.1016/j.jaci.2013.07.026 103. Kappeler L, Filho CDM, Dupont J et al (2008) Brain IGF-1 re- 119. Salinas M, López-Valdaliso R, Martín D et al (2000) Inhibition of ceptors control mammalian growth and lifespan through a neuro- PKB/Akt1 by C2-ceramide involves activation of ceramide- endocrine mechanism. PLoS Biol 6:e254. https://doi.org/10.1371/ activated protein phosphatase in PC12 cells. Mol Cell Neurosci journal.pbio.0060254 15:156–169. https://doi.org/10.1006/mcne.1999.0813 104. Taguchi A, Wartschow LM, White MF (2007) Brain IRS2 signal- 120. Qin J, Berdyshev E, Poirer C et al (2012) Neutral ing coordinates life span and nutrient homeostasis. Science sphingomyelinase 2 deficiency increases hyaluronan synthesis 317(80):369–372. https://doi.org/10.1126/science.1142179 by up-regulation of Hyaluronan synthase 2 through decreased 105. Bokov AF, Garg N, Ikeno Y et al (2011) Does reduced IGF-1R ceramide production and activation of Akt. J Biol Chem 287: signaling in Igf1r+/- mice alter aging? PLoS One 6:e26891. 13620–13632. https://doi.org/10.1074/jbc.M111.304857 https://doi.org/10.1371/journal.pone.0026891 121. Gómez-Muñoz A, Kong JY, Parhar K et al (2005) Ceramide-1- 106. Deelen J, Uh H-W, Monajemi R et al (2013) Gene set analysis of phosphate promotes cell survival through activation of the phos- GWAS data for human longevity highlights the relevance of the phatidylinositol 3-kinase/protein kinase B pathway. FEBS Lett insulin/IGF-1 signaling and telomere maintenance pathways. Age 579:3744–3750. https://doi.org/10.1016/j.febslet.2005.05.067 (Dordr) 35:235–249. https://doi.org/10.1007/s11357-011-9340-3 122. Gangoiti P, Granado MH, Wang SW et al (2008) Ceramide 1- 107. Suh Y, Atzmon G, Cho M-O et al (2008) Functionally significant phosphate stimulates macrophage proliferation through activation insulin-like growth factor I receptor mutations in centenarians. of the PI3-kinase/PKB, JNK and ERK1/2 pathways. Cell Signal Proc Natl Acad Sci U S A 105:3438–3442. https://doi.org/10. 20:726–736. https://doi.org/10.1016/j.cellsig.2007.12.008 1073/pnas.0705467105 123. Huwiler A, Kotelevets N, Xin C et al (2011) Loss of sphingosine 108. Ceda GP, Dall’Aglio E, Maggio M et al (2005) Clinical implica- kinase-1 in carcinoma cells increases formation of reactive oxygen tions of the reduced activity of the GH-IGF-I axis in older men. J species and sensitivity to doxorubicin-induced DNA damage. Br J Endocrinol Investig 28:96–100 Pharmacol 162:532–543. https://doi.org/10.1111/j.1476-5381. 109. Okereke O, Kang JH, Ma J et al (2007) Plasma IGF-I levels and 2010.01053.x cognitive performance in older women. Neurobiol Aging 28:135– 124. Dbaibo GS, Pushkareva MY, Rachid RA et al (1998) p53- 142. https://doi.org/10.1016/j.neurobiolaging.2005.10.012 dependent ceramide response to genotoxic stress. J Clin Invest 110. Morel GR, León ML, Uriarte M et al (2017) Therapeutic potential 102:329–339. https://doi.org/10.1172/JCI1180 of IGF-I on hippocampal neurogenesis and function during aging. 125. Cutler RG, Pedersen WA, Camandola S et al (2002) Evidence that Neurogenesis 4:e1259709. https://doi.org/10.1080/23262133. accumulation of ceramides and cholesterol esters mediates oxidative 2016.1259709 stress-induced death of motor neurons in amyotrophic lateral scle- 111. Pardo J, Abba MC, Lacunza E et al (2018) IGF-I gene therapy in rosis. Ann Neurol 52:448–457. https://doi.org/10.1002/ana.10312 aging rats modulates hippocampal genes relevant to memory func- 126. Liu B, Hannun YA (1997) Inhibition of the neutral magnesium- tion. J Gerontol A Biol Sci Med Sci 73:459–467. https://doi.org/ dependent sphingomyelinase by glutathione. J Biol Chem 272: 10.1093/gerona/glx125 16281–16287 112. Bernacchioni C, Cencetti F, Blescia S et al (2012) Sphingosine 127. Kosicek M, Zetterberg H, Andreasen N et al (2012) Elevated kinase/sphingosine 1-phosphate axis: a new player for insulin- cerebrospinal fluid sphingomyelin levels in prodromal like growth factor-1-induced myoblast differentiation. Skelet Alzheimer’s disease. Neurosci Lett 516:302–305. https://doi.org/ Muscle 2:15. https://doi.org/10.1186/2044-5040-2-15 10.1016/j.neulet.2012.04.019 113. Marfe G, Di Stefano C, Gambacurta A et al (2011) Sphingosine 128. Olsen ASB, Færgeman NJ (2017) Sphingolipids: membrane mi- kinase 1 overexpression is regulated by signaling through PI3K, crodomains in brain development, function and neurological dis- AKT2, and mTOR in imatinib-resistant chronic myeloid leukemia eases. Open Biol 7:170069. https://doi.org/10.1098/rsob.170069 cells. Exp Hematol 39:653–665.e6. https://doi.org/10.1016/j. 129. Ohno-Iwashita Y, Shimada Y, Hayashi M, Inomata M (2010) exphem.2011.02.013 Plasma membrane microdomains in aging and disease. Geriatr 114. Till KJ, Pettitt AR, Slupsky JR (2015) Expression of functional Gerontol Int 10(Suppl 1):S41–S52. https://doi.org/10.1111/j. sphingosine-1 phosphate receptor-1 is reduced by B cell receptor 1447-0594.2010.00600.x signaling and increased by inhibition of PI3 kinase δ but not SYK 130. Badawy SMM, Okada T, Kajimoto T et al (2018) Extracellular α- or BTK in chronic lymphocytic leukemia cells. J Immunol 194: synuclein drives sphingosine 1-phosphate receptor subtype 1 out of 2439–2446. https://doi.org/10.4049/jimmunol.1402304 lipid rafts, leading to impaired inhibitory G-protein signaling. J Biol 115. Giussani P, Brioschi L, Bassi R et al (2009) Phosphatidylinositol Chem 293:8208–8216. https://doi.org/10.1074/jbc.RA118.001986 3-kinase/AKT pathway regulates the endoplasmic reticulum to 131. Grimm MOW, Grösgen S, Rothhaar TL et al (2011) Intracellular golgi traffic of ceramide in glioma cells: a link between lipid APP domain regulates serine-palmitoyl-CoA transferase Mol Neurobiol (2019) 56:5436–5455 5451

expression and is affected in Alzheimer’s disease. Int J Alzheimers 148. Cutler RG, Kelly J, Storie K et al (2004) Involvement of oxidative Dis 2011:695413. https://doi.org/10.4061/2011/695413 stress-induced abnormalities in ceramide and cholesterol metabo- 132. Kang MS, Baek S-H, Chun YS et al (2013) Modulation of lipid lism in brain aging and Alzheimer’s disease. Proc Natl Acad Sci U kinase PI4KIIα activity and lipid raft association of presenilin 1 S A 101:2070–2075. https://doi.org/10.1073/pnas.0305799101 underlies γ-secretase inhibition by ginsenoside (20S)-Rg3. J Biol 149. Katsel P, Li C, Haroutunian V (2007) Gene expression alterations Chem 288:20868–20882. https://doi.org/10.1074/jbc.M112.445734 in the sphingolipid metabolism pathways during progression of 133. Amaro M, Šachl R, Aydogan G et al (2016) GM1 ganglioside dementia and Alzheimer’s disease: a shift toward ceramide accu- inhibits β-amyloid oligomerization induced by sphingomyelin. mulation at the earliest recognizable stages of Alzheimer’sdis- Angew Chem Int Ed Engl 55:9411–9415. https://doi.org/10. ease? Neurochem Res 32:845–856. https://doi.org/10.1007/ 1002/anie.201603178 s11064-007-9297-x 134. Rondelli V, Brocca P, Motta S et al (2016) Amyloid-β peptides in 150. He X, Huang Y, Li B et al (2010) Deregulation of sphingolipid interaction with raft-mime model membranes: a neutron reflectiv- metabolism in Alzheimer’s disease. Neurobiol Aging 31:398–408. ity insight. Sci Rep 6:20997. https://doi.org/10.1038/srep20997 https://doi.org/10.1016/j.neurobiolaging.2008.05.010 135. Peters I, Igbavboa U, Schütt T et al (2009) The interaction of beta- 151. Bandaru VVR, Troncoso J, Wheeler D et al (2009) ApoE4 dis- amyloid protein with cellular membranes stimulates its own pro- rupts sterol and sphingolipid metabolism in Alzheimer’s but not duction. Biochim Biophys Acta 1788:964–972. https://doi.org/10. normal brain. Neurobiol Aging 30:591–599. https://doi.org/10. 1016/j.bbamem.2009.01.012 1016/j.neurobiolaging.2007.07.024 136. Sawamura N, Ko M, Yu W et al (2004) Modulation of amyloid 152. Lee J-T, Xu J, Lee J-M et al (2004) Amyloid-beta peptide induces precursor protein cleavage by cellular sphingolipids. J Biol Chem oligodendrocyte death by activating the neutral 279:11984–11991. https://doi.org/10.1074/jbc.M309832200 sphingomyelinase-ceramide pathway. J Cell Biol 164:123–131. 137. Puglielli L, Ellis BC, Saunders AJ, Kovacs DM (2003) Ceramide https://doi.org/10.1083/jcb.200307017 stabilizes beta-site amyloid precursor protein-cleaving enzyme 1 153. Jana A, Pahan K (2004) Fibrillar amyloid-beta peptides kill human and promotes amyloid beta-peptide biogenesis. J Biol Chem 278: primary neurons via NADPH oxidase-mediated activation of neutral 19777–19783. https://doi.org/10.1074/jbc.M300466200 sphingomyelinase. Implications for Alzheimer’s disease J Biol 138. Takasugi N, Sasaki T, Shinohara M et al (2015) Synthetic cer- Chem 279:51451–51459. https://doi.org/10.1074/jbc.M404635200 amide analogues increase amyloid-β 42 production by modulating 154. Gomez-Brouchet A, Pchejetski D, Brizuela L et al (2007) Critical γ-secretase activity. Biochem Biophys Res Commun 457:194– role for sphingosine kinase-1 in regulating survival of neuroblas- 199. https://doi.org/10.1016/j.bbrc.2014.12.087 toma cells exposed to amyloid-beta peptide. Mol Pharmacol 72: 139. Asle-Rousta M, Kolahdooz Z, Oryan S et al (2013) FTY720 341–349. https://doi.org/10.1124/mol.106.033738 (fingolimod) attenuates beta-amyloid peptide (Aβ42)-induced im- 155. Ayasolla K, Khan M, Singh AK, Singh I (2004) Inflammatory me- pairment of spatial learning and memory in rats. J Mol Neurosci diator and beta-amyloid (25-35)-induced ceramide generation and 50:524–532. https://doi.org/10.1007/s12031-013-9979-6 iNOS expression are inhibited by vitamin E. Free Radic Biol Med 140. Joshi P, Gabrielli M, Ponzoni L et al (2017) Fingolimod limits 37:325–338. https://doi.org/10.1016/j.freeradbiomed.2004.04.007 acute Aβ neurotoxicity and promotes synaptic versus 156. Zeng C, Lee JT, Chen H et al (2005) Amyloid-beta peptide enhances extrasynaptic NMDA receptor functionality in hippocampal neu- tumor necrosis factor-alpha-induced iNOS through neutral rons. Sci Rep 7:41734. https://doi.org/10.1038/srep41734 sphingomyelinase/ceramide pathway in oligodendrocytes. J 141. Takasugi N, Sasaki T, Ebinuma I et al (2013) FTY720/fingolimod, a Neurochem 94:703–712. https://doi.org/10.1111/j.1471-4159. sphingosine analogue, reduces amyloid-β production in neurons. 2005.03217.x PLoS One 8:e64050. https://doi.org/10.1371/journal.pone.0064050 157. Martinez TN, Chen X, Bandyopadhyay S et al (2012) Ceramide 142. Takasugi N, Sasaki T, Suzuki K et al (2011) BACE1 activity is sphingolipid signaling mediates tumor necrosis factor (TNF)-de- modulated by cell-associated sphingosine-1-phosphate. J pendent toxicity via caspase signaling in dopaminergic neurons. Neurosci 31:6850–6857. https://doi.org/10.1523/JNEUROSCI. Mol Neurodegener 7:45. https://doi.org/10.1186/1750-1326-7-45 6467-10.2011 158. Alessenko AV, Bugrova AE, Dudnik LB (2004) Connection of 143. Dominguez G, Maddelein M-L, Pucelle M et al (2018) Neuronal lipid peroxide oxidation with the sphingomyelin pathway in the sphingosine kinase 2 subcellular localization is altered in development of Alzheimer’s disease. Biochem Soc Trans 32:144– Alzheimer’s disease brain. Acta Neuropathol Commun 6:25. 146. https://doi.org/10.1000/182 https://doi.org/10.1186/s40478-018-0527-z 159. Grimm MOW, Grimm HS, Pätzold AJ et al (2005) Regulation of 144. Ceccom J, Loukh N, Lauwers-Cances V et al (2014) Reduced cholesterol and sphingomyelin metabolism by amyloid-β and sphingosine kinase-1 and enhanced sphingosine 1-phosphate ly- presenilin. Nat Cell Biol 7:1118–1123. https://doi.org/10.1038/ ase expression demonstrate deregulated sphingosine 1-phosphate ncb1313 signaling in Alzheimer’s disease. Acta Neuropathol Commun 2: 160. Dinkins MB, Enasko J, Hernandez C et al (2016) Neutral 12. https://doi.org/10.1186/2051-5960-2-12 sphingomyelinase-2 deficiency ameliorates Alzheimer’sdisease 145. Karaca I, Tamboli IY, Glebov K et al (2014) Deficiency of pathology and improves cognition in the 5XFAD mouse. J sphingosine-1-phosphate lyase impairs lysosomal metabolism of Neurosci 36:8653–8667. https://doi.org/10.1523/JNEUROSCI. the amyloid precursor protein. J Biol Chem 289:16761–16772. 1429-16.2016 https://doi.org/10.1074/jbc.M113.535500 161. Geekiyanage H, Upadhye A, Chan C (2013) Inhibition of serine 146. Jesko H, Okada T, Strosznajder RP, Nakamura S (2014) palmitoyltransferase reduces Aβ and tau hyperphosphorylation in Sphingosine kinases modulate the secretion of amyloid β precur- a murine model: a safe therapeutic strategy for Alzheimer’sdis- sor protein from SH-SY5Y neuroblastoma cells: the role of α- ease. Neurobiol Aging 34:2037–2051. https://doi.org/10.1016/j. synuclein. Folia Neuropathol 52:70–78 neurobiolaging.2013.02.001 147. Jęśko H, Wencel PL, Lukiw WJ, Strosznajder RP (2018) 162. Panchal M, Gaudin M, Lazar AN et al (2014) Ceramides and Modulatory effects of fingolimod (FTY720) on the expression sphingomyelinases in senile plaques. Neurobiol Dis 65:193–201. of sphingolipid metabolism-related genes in an animal model of https://doi.org/10.1016/j.nbd.2014.01.010 Alzheimer’s disease. Mol Neurobiol https://doi.org/10.1007/ 163. Gassowska M, Cieslik M, Wilkaniec A, Strosznajder JB (2014) s12035-018-1040-x Sphingosine kinases/sphingosine-1-phosphate and death 5452 Mol Neurobiol (2019) 56:5436–5455

Signalling in APP-transfected cells. Neurochem Res 39:645–652. 180. Ngolab J, Trinh I, Rockenstein E et al (2017) Brain-derived https://doi.org/10.1007/s11064-014-1240-3 exosomes from dementia with Lewy bodies propagate α- 164. Kuhle J, Disanto G, Lorscheider J et al (2015) Fingolimod and synuclein pathology. Acta Neuropathol Commun 5:46. https:// CSF neurofilament light chain levels in relapsing-remitting multi- doi.org/10.1186/s40478-017-0445-5 ple sclerosis. Neurology 84:1639–1643. https://doi.org/10.1212/ 181. Wang Y,Balaji V,Kaniyappan S et al (2017) The release and trans- WNL.0000000000001491 synaptic transmission of tau via exosomes. Mol Neurodegener 12: 165. Xia P, Gamble JR, Rye KA et al (1998) Tumor necrosis factor- 5. https://doi.org/10.1186/s13024-016-0143-y alpha induces adhesion molecule expression through the sphingo- 182. Kajimoto T, Mohamed NNI, Badawy SMM et al (2018) sine kinase pathway. Proc Natl Acad Sci U S A 95:14196–14201 Involvement of Gβγ subunits of Gi protein coupled with S1P 166. Kaneider NC, Lindner J, Feistritzer C et al (2004) The immune mod- receptor on multivesicular endosomes in F-actin formation and ulator FTY720 targets sphingosine-kinase-dependent migration of hu- cargo sorting into exosomes. J Biol Chem 293:245–253. https:// man monocytes in response to amyloid beta-protein and its precursor. doi.org/10.1074/jbc.M117.808733 FASEB J 18:1309–1311. https://doi.org/10.1096/fj.03-1050fje 183. Dinkins MB, Dasgupta S, Wang G et al (2014) Exosome reduction 167. Dinkins MB, Dasgupta S, Wang G et al (2015) The 5XFAD mouse in vivo is associated with lower amyloid plaque load in the 5XFAD model of Alzheimer’s disease exhibits an age-dependent increase in mouse model of Alzheimer’s disease. Neurobiol Aging 35:1792– anti-ceramide IgG and exogenous administration of ceramide fur- 1800. https://doi.org/10.1016/j.neurobiolaging.2014.02.012 ther increases anti-ceramide titers and amyloid plaque burden. J 184. Broe M, Shepherd CE, Mann DMA et al (2005) Insoluble alpha- Alzheimers Dis 46:55–61. https://doi.org/10.3233/JAD-150088 synuclein in Alzheimer’s disease without Lewy body formation. 168. Liu C-C, Kanekiyo T, Xu H et al (2013) Apolipoprotein E and Neurotox Res 7:69–76 Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol 185. Emmanouilidou E, Vekrellis K (2016) Exocytosis and spreading 9:106–118. https://doi.org/10.1038/nrneurol.2012.263 of normal and aberrant α-synuclein. Brain Pathol 26:398–403. 169. Kurano M, Tsukamoto K, Hara M et al (2015) LDL receptor and https://doi.org/10.1111/bpa.12373 ApoE are involved in the clearance of ApoM-associated sphingo- 186. Christensen DP, Ejlerskov P, Rasmussen I, Vilhardt F (2016) sine 1-phosphate. J Biol Chem 290:2477–2488. https://doi.org/10. Reciprocal signals between microglia and neurons regulate α- 1074/jbc.M114.596445 synuclein secretion by exophagy through a neuronal cJUN-N- 170. Han X (2010) Multi-dimensional mass spectrometry-based shot- terminal kinase-signaling axis. J Neuroinflammation 13:59. gun lipidomics and the altered lipids at the mild cognitive impair- https://doi.org/10.1186/s12974-016-0519-5 ment stage of Alzheimer’s disease. Biochim Biophys Acta 1801: 187. Paillusson S, Clairembault T, Biraud M et al (2013) Activity- 774–783. https://doi.org/10.1016/j.bbalip.2010.01.010 dependent secretion of alpha-synuclein by enteric neurons. J 171. Gizaw ST, Ohashi T, Tanaka M et al (2016) Glycoblotting method Neurochem 125:512–517. https://doi.org/10.1111/jnc.12131 allows for rapid and efficient glycome profiling of human 188. Ejlerskov P, Rasmussen I, Nielsen TT et al (2013) Tubulin Alzheimer’s disease brain, serum and cerebrospinal fluid towards polymerization-promoting protein (TPPP/p25α)promotesuncon- potential biomarker discovery. Biochim Biophys Acta 1860: ventional secretion of α-synuclein through exophagy by 1716–1727. https://doi.org/10.1016/j.bbagen.2016.03.009 impairing autophagosome-lysosome fusion. J Biol Chem 288: 172. Mielke MM, Lyketsos CG (2010) Alterations of the sphingolipid 17313–17335. https://doi.org/10.1074/jbc.M112.401174 pathway in Alzheimer’s disease: new biomarkers and treatment 189. Hasegawa T, Konno M, Baba T et al (2011) The AAA-ATPase targets? NeuroMolecular Med 12:331–340. https://doi.org/10. VPS4 regulates extracellular secretion and lysosomal targeting of 1007/s12017-010-8121-y α-synuclein. PLoS One 6:e29460. https://doi.org/10.1371/journal. 173. Yuyama K, Mitsutake S, Igarashi Y (2014) Pathological roles of pone.0029460 ceramide and its metabolites in metabolic syndrome and 190. Papadopoulos VE, Nikolopoulou G, Antoniadou I et al (2018) Alzheimer’s disease. Biochim Biophys Acta 1841:793–798. Modulation of β-glucocerebrosidase increases α-synuclein secretion https://doi.org/10.1016/j.bbalip.2013.08.002 and exosome release in mouse models of Parkinson’s disease. Hum 174. Wang G, Dinkins M, He Q et al (2012) Astrocytes secrete Mol Genet 27:1696–1710. https://doi.org/10.1093/hmg/ddy075 exosomes enriched with proapoptotic ceramide and prostate apo- 191. Zhang S, Eitan E, Wu T-Y, Mattson MP (2018) Intercellular transfer ptosis response 4 (PAR-4): potential mechanism of apoptosis in- of pathogenic α-synuclein by extracellular vesicles is induced by the duction in Alzheimer disease (AD). J Biol Chem 287:21384– lipid peroxidation product 4-hydroxynonenal. Neurobiol Aging 61: 21395. https://doi.org/10.1074/jbc.M112.340513 52–65. https://doi.org/10.1016/j.neurobiolaging.2017.09.016 175. Sharples RA, Vella LJ, Nisbet RM et al (2008) Inhibition of 192. Abbott SK, Li H, Muñoz SS et al (2014) Altered ceramide acyl gamma-secretase causes increased secretion of amyloid precursor chain length and ceramide synthase gene expression in protein C-terminal fragments in association with exosomes. Parkinson’s disease. Mov Disord 29:518–526 FASEB J 22:1469–1478. https://doi.org/10.1096/fj.07-9357.com 193. Sivasubramanian M, Kanagaraj N, Dheen ST, Tay SSW (2015) 176. Rajendran L, Honsho M, Zahn TR et al (2006) Alzheimer’sdis- Sphingosine kinase 2 and sphingosine-1-phosphate promotes mi- ease beta-amyloid peptides are released in association with tochondrial function in dopaminergic neurons of mouse model of exosomes. Proc Natl Acad Sci 103:11172–11177. https://doi.org/ Parkinson’s disease and in MPP+-treated MN9D cells in vitro. 10.1073/pnas.0603838103 Neuroscience 290:636–648. https://doi.org/10.1016/j. 177. Dinkins MB, Wang G, Bieberich E (2017) Sphingolipid-enriched neuroscience.2015.01.032 extracellular vesicles and Alzheimer’s disease: a decade of research. 194. Motyl J, Przykaza Ł, Boguszewski PM, Kosson P, Strosznajder JB J Alzheimers Dis 60:757–768. https://doi.org/10.3233/JAD-160567 (2018) Pramipexole and Fingolimod exert neuroprotection in a 178. Yuyama K, Sun H, Mitsutake S, Igarashi Y (2012) Sphingolipid- mouse model of Parkinson's disease by activation of sphingosine modulated exosome secretion promotes clearance of amyloid-β kinase 1 and Akt kinase. Neuropharmacology 135:139–150. by microglia. J Biol Chem 287:10977–10989. https://doi.org/10. https://doi.org/10.1016/j.neuropharm.2018.02.023 1074/jbc.M111.324616 195. Ferrazza R, Cogo S, Melrose H et al (2016) LRRK2 deficiency 179. Yuyama K, Igarashi Y (2017) Exosomes as carriers of impacts ceramide metabolism in brain. Biochem Biophys Res Alzheimer’s amyloid-ß. Front Neurosci 11(229). https://doi.org/ Commun 478:1141–1146. https://doi.org/10.1016/j.bbrc.2016. 10.3389/fnins.2017.00229 08.082 Mol Neurobiol (2019) 56:5436–5455 5453

196. Bras J, Singleton A, Cookson MR, Hardy J (2008) Emerging 213. Sato S, Li Y,Hattori N (2017) Lysosomal defects in ATP13A2 and pathways in genetic Parkinson’s disease: potential role of cer- GBA associated familial Parkinson’s disease. J Neural Transm amide metabolism in Lewy body disease. FEBS J 275:5767– 124:1395–1400. https://doi.org/10.1007/s00702-017-1779-7 5773. https://doi.org/10.1111/j.1742-4658.2008.06709.x 214. Pchelina S, Emelyanov A, Baydakova G et al (2017) Oligomeric 197. Brodowicz J, Przegaliński E, Müller CP, Filip M (2018) Ceramide α-synuclein and glucocerebrosidase activity levels in GBA- and its related neurochemical networks as targets for some brain associated Parkinson’s disease. Neurosci Lett 636:70–76. https:// disorder therapies. Neurotox Res 33:474–484. https://doi.org/10. doi.org/10.1016/j.neulet.2016.10.039 1007/s12640-017-9798-6 215. Jin H, Chen J, Li K et al (2018) A novel p.L216I mutation in the 198. Mielke MM, Maetzler W, Haughey NJ et al (2013) Plasma cer- glucocerebrosidase gene is associated with Parkinson’sdiseasein amide and glucosylceramide metabolism is altered in sporadic Han Chinese patients. Neurosci Lett 674:66–69. https://doi.org/ Parkinson’s disease and associated with cognitive impairment: a 10.1016/j.neulet.2018.03.017 pilot study. PLoS One 8:e73094. https://doi.org/10.1371/journal. 216. Arkadir D, Dinur T, Mullin S et al (2018) Trio approach reveals higher pone.0073094 risk of PD in carriers of severe vs. mild GBA mutations. Blood Cells 199. Xing Y, Tang Y, Zhao L et al (2016) Associations between plasma Mol Dis 68:115–116. https://doi.org/10.1016/j.bcmd.2016.11.007 ceramides and cognitive and neuropsychiatric manifestations in 217. Jesús S, Huertas I, Bernal-Bernal I et al (2016) GBA variants influ- Parkinson’s disease dementia. J Neurol Sci 370:82–87. https:// ence motor and non-motor features of Parkinson’s disease. PLoS doi.org/10.1016/j.jns.2016.09.028 One 11:e0167749. https://doi.org/10.1371/journal.pone.0167749 200. MaoC,YangJ,WangHetal(2017)SMPD1variantsinChinese 218. Swan M, Doan N, Ortega RA et al (2016) Neuropsychiatric char- ’ Han patients with sporadic Parkinson s disease. Parkinsonism Relat acteristics of GBA-associated Parkinson disease. J Neurol Sci 370: – Disord 34:59 61. https://doi.org/10.1016/j.parkreldis.2016.10.014 63–69. https://doi.org/10.1016/j.jns.2016.08.059 201. Gan-Or Z, Orr-Urtreger A, Alcalay RN et al (2015) The emerging 219. Parnetti L, Paciotti S, Eusebi P et al (2017) Cerebrospinal fluid β- ’ role of SMPD1 mutations in Parkinson s disease: implications for glucocerebrosidase activity is reduced in parkinson’s disease patients. – future studies. Parkinsonism Relat Disord 21:1294 1295. https:// Mov Disord 32:1423–1431. https://doi.org/10.1002/mds.27136 doi.org/10.1016/j.parkreldis.2015.08.018 220. Lunde KA, Chung J, Dalen I et al (2018) Association of 202. Dagan E, Schlesinger I, Ayoub M et al (2015) The contribution of glucocerebrosidase polymorphisms and mutations with dementia Niemann-Pick SMPD1 mutations to Parkinson disease in ’ – – in incident Parkinson s disease. Alzheimers Dement 14:1293 Ashkenazi Jews. Parkinsonism Relat Disord 21:1067 1071. 1301. https://doi.org/10.1016/j.jalz.2018.04.006 https://doi.org/10.1016/j.parkreldis.2015.06.016 221. Taguchi YV, Liu J, Ruan J et al (2017) Glucosylsphingosine pro- 203. Foo J-N, Liany H, Bei J-X et al (2013) Rare lysosomal enzyme motes α-synuclein pathology in mutant GBA-associated gene SMPD1 variant (p.R591C) associates with Parkinson’sdis- Parkinson’s disease. J Neurosci 37:9617–9631. https://doi.org/ ease. Neurobiol Aging 34:2890.e13–2890.e15. https://doi.org/10. 10.1523/JNEUROSCI.1525-17.2017 1016/j.neurobiolaging.2013.06.010 222. Guedes LC, Chan RB, Gomes MA et al (2017) Serum lipid alter- 204. Deng S, Deng X, Song Z et al (2016) Systematic genetic analysis of the ations in GBA-associated Parkinson’s disease. Parkinsonism Relat SMPD1 gene in Chinese patients with Parkinson’sdisease.Mol Disord 44:58–65. https://doi.org/10.1016/j.parkreldis.2017.08.026 Neurobiol 53:5025–5029. https://doi.org/10.1007/s12035-015-9426-5 205. Pyszko J, Strosznajder JB (2014) Sphingosine kinase 1 and 223. Mazzulli JR, Xu Y-H, Sun Y et al (2011) Gaucher disease glucocerebrosidase and α-synuclein form a bidirectional patho- sphingosine-1-phosphate in oxidative stress evoked by 1- – methyl-4-Phenylpyridinium (MPP+) in human dopaminergic neu- genic loop in synucleinopathies. Cell 146:37 52. https://doi.org/ ronal cells. Mol Neurobiol 50:38–48. https://doi.org/10.1007/ 10.1016/j.cell.2011.06.001 s12035-013-8622-4 224. Abul Khair SB, Dhanushkodi NR, Ardah MT et al (2018) 206. Mitroi DN, Karunakaran I, Gräler M et al (2017) SGPL1 (sphin- Silencing of glucocerebrosidase gene in Drosophila enhances ’ α gosine phosphate lyase 1) modulates neuronal autophagy via the aggregation of Parkinson s disease associated -synuclein phosphatidylethanolamine production. Autophagy 13:885–899. mutant A53T and affects locomotor activity. Front Neurosci 12: https://doi.org/10.1080/15548627.2017.1291471 81. https://doi.org/10.3389/fnins.2018.00081 207. Ren M, Han M, Wei X et al (2017) FTY720 attenuates 6-OHDA- 225. Jung O, Patnaik S, Marugan J et al (2016) Progress and potential associated dopaminergic degeneration in cellular and mouse par- of non-inhibitory small molecule chaperones for the treatment of kinsonian models. Neurochem Res 42:686–696. https://doi.org/ Gaucher disease and its implications for Parkinson disease. Expert – 10.1007/s11064-016-2125-4 Rev Proteomics 13:471 479. https://doi.org/10.1080/14789450. 208. Zhao P, Yang X, Yang L et al (2017) Neuroprotective effects of 2016.1174583 fingolimod in mouse models of Parkinson’sdisease.FASEBJ31: 226. Di Pardo A, Amico E, Basit A et al (2017) Defective sphingosine-1- 172–179. https://doi.org/10.1096/fj.201600751R phosphate metabolism is a druggable target in Huntington’s disease. 209. Komnig D, Dagli CT, Habib P, et al (2018) Fingolimod (FTY720) Sci Rep 7:5280. https://doi.org/10.1038/s41598-017-05709-y is not protective in the subacute MPTP mouse model of Parkinson 227. Di Pardo A, Basit A, Armirotti A et al (2017) De novo synthesis of disease and does not lead to a sustainable increase of brain BDNF. sphingolipids is defective in experimental models of Huntington’sdis- JNeurochemhttps://doi.org/10.1111/jnc.14575 ease. Front Neurosci 11:698. https://doi.org/10.3389/fnins.2017.00698 210. Kim WS, Halliday GM (2012) Changes in sphingomyelin level 228. Hunter M, Demarais NJ, Faull RLM et al (2018) Subventricular affect alpha-synuclein and ABCA5 expression. J Parkinsons Dis zone lipidomic architecture loss in Huntington’s disease. J 2:41–46. https://doi.org/10.3233/JPD-2012-11059 Neurochem 146:613–630. https://doi.org/10.1111/jnc.14468 211. Bienias K, Fiedorowicz A, Sadowska A et al (2016) Regulation of 229. Pirhaji L, Milani P, Leidl M et al (2016) Revealing disease- sphingomyelin metabolism. Pharmacol Rep 68:570–581. https:// associated pathways by network integration of untargeted meta- doi.org/10.1016/j.pharep.2015.12.008 bolomics. Nat Methods 13:770–776. https://doi.org/10.1038/ 212. Vidal-Martínez G, Vargas-Medrano J, Gil-Tommee C et al (2016) nmeth.3940 FTY720/fingolimod reduces synucleinopathy and improves gut 230. Pirhaji L, Milani P, Dalin S et al (2017) Identifying therapeutic motility in A53T mice: contributions of pro-brain-derived neuro- targets by combining transcriptional data with ordinal clinical trophic factor (PRO-BDNF) and mature BDNF. J Biol Chem 291: measurements. Nat Commun 8:623. https://doi.org/10.1038/ 20811–20821. https://doi.org/10.1074/jbc.M116.744029 s41467-017-00353-6 5454 Mol Neurobiol (2019) 56:5436–5455

231. Mangiarini L, Sathasivam K, Seller M et al (1996) Exon 1 of the HD 247. Gui Y, Liu H, Zhang L et al (2015) Altered microRNA profiles in gene with an expanded CAG repeat is sufficient to cause a progres- cerebrospinal fluid exosome in Parkinson disease and Alzheimer sive neurological phenotype in transgenic mice. Cell 87:493–506 disease. Oncotarget 6:37043–37053. https://doi.org/10.18632/ 232. Di Pardo A, Amico E, Favellato M et al (2014) FTY720 oncotarget.6158 (fingolimod) is a neuroprotective and disease-modifying agent 248. Xin Q, Li J, Dang J et al (2015) miR-155 deficiency ameliorates in cellular and mouse models of Huntington disease. Hum Mol autoimmune inflammation of systemic lupus erythematosus by Genet 23:2251–2265. https://doi.org/10.1093/hmg/ddt615 targeting S1pr1 in Faslpr/lpr mice. J Immunol 194:5437–5445. 233. Di Pardo A, Castaldo S, Amico E et al (2018) Stimulation of https://doi.org/10.4049/jimmunol.1403028 S1PR5 with A-971432, a selective agonist, preserves blood- 249. Jaber V, Zhao Y, Lukiw WJ (2017) Alterations in micro RNA- brain barrier integrity and exerts therapeutic effect in an animal messenger RNA (miRNA-mRNA) coupled signaling networks in model of Huntington’s disease. Hum Mol Genet 27:2490–2501. sporadic Alzheimer’s disease (AD) hippocampal CA1. J https://doi.org/10.1093/hmg/ddy153 Alzheimer’sDisPark7:https://doi.org/10.4172/2161-0460.1000312 234. Miguez A, García-Díaz Barriga G, Brito V, Straccia M, Giralt A, 250. Pogue AI, Lukiw WJ (2018) Up-regulated pro-inflammatory Ginés S, Canals JM, Alberch J (2015) Fingolimod (FTY720) en- microRNAs (miRNAs) in Alzheimer’s disease (AD) and age- hances hippocampal synaptic plasticity and memory in related macular degeneration (AMD). Cell Mol Neurobiol 38: Huntington’s disease by preventing p75NTR up-regulation and 1021–1031. https://doi.org/10.1007/s10571-017-0572-3 astrocyte-mediated inflammation. Hum Mol Genet 24(17):4958– 251. MIR155 Gene—GeneCards | MIR155 RNA Gene. https://www. 4970. https://doi.org/10.1093/hmg/ddv218 .org/cgi-bin/carddisp.pl?gene=MIR155. Accessed 23 235. Di Pardo A, Maglione V (2018) The S1P axis: new exciting route Jul 2018 for treating Huntington’s disease. Trends Pharmacol Sci 39:468– 252. Murphy EJ, Schapiro MB, Rapoport SI, Shetty HU (2000) 480. https://doi.org/10.1016/j.tips.2018.02.009 Phospholipid composition and levels are altered in Down syn- – 236. Henriques A, Croixmarie V, Priestman DA et al (2015) drome brain. Brain Res 867:9 18 Amyotrophic lateral sclerosis and denervation alter sphingolipids 253. Zhao Y, Jaber V, Percy ME, Lukiw WJ (2017) A microRNA and up-regulate glucosylceramide synthase. Hum Mol Genet 24: cluster (let-7c, miRNA-99a, miRNA-125b, miRNA-155 and 7390–7405. https://doi.org/10.1093/hmg/ddv439 miRNA-802) encoded at chr21q21.1-chr21q21.3 and the pheno- ’ 237. Dodge JC, Treleaven CM, Pacheco J et al (2015) typic diversity of Down s syndrome (DS; trisomy 21). J Nat Sci 3: Glycosphingolipids are modulators of disease pathogenesis in 254. Charkiewicz K, Blachnio-Zabielska A, Zbucka-Kretowska M et al amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A 112: (2015) Maternal plasma and amniotic fluid sphingolipids profiling 8100–8105. https://doi.org/10.1073/pnas.1508767112 in fetal Down syndrome. PLoS One 10:e0127732. https://doi.org/ 10.1371/journal.pone.0127732 238. Henriques A, Croixmarie V, Bouscary A, Mosbach A, Keime C, 255. Zhao Y, Lukiw WJ (2018) Bacteroidetes neurotoxins and inflam- Boursier-Neyret C, Walter B, Spedding M et al (2018) matory neurodegeneration. Mol Neurobiol 55:9100–9107. https:// Sphingolipid metabolism is dysregulated at transcriptomic and doi.org/10.1007/s12035-018-1015-y metabolic levels in the spinal cord of an animal model of amyo- 256. Heaver SL, Johnson EL, Ley RE (2018) Sphingolipids in host- trophic lateral sclerosis. Front Mol Neurosci 10:433. https://doi. microbial interactions. Curr Opin Microbiol 43:92–99. https://doi. org/10.3389/fnmol.2017.00433 org/10.1016/j.mib.2017.12.011 239. Potenza RL, De Simone R, Armida M et al (2016) Fingolimod: a 257. Hannun YA, Obeid LM (2008) Principles of bioactive lipid sig- disease-modifier drug in a mouse model of amyotrophic lateral nalling: lessons from sphingolipids. Nat Rev Mol Cell Biol 9:139– sclerosis. Neurotherapeutics 13:918–927. https://doi.org/10.1007/ 150. https://doi.org/10.1038/nrm2329 s13311-016-0462-2 258. Packer AN, Xing Y, Harper SQ et al (2008) The bifunctional 240. Batistela MS, Josviak ND, Sulzbach CD, de Souza RLR (2017) microRNA miR-9/miR-9* regulates REST and CoREST and is An overview of circulating cell-free microRNAs as putative bio- downregulated in Huntington’s disease. J Neurosci 28:14341– markers in Alzheimer’s and Parkinson’s diseases. Int J Neurosci – 14346. https://doi.org/10.1523/JNEUROSCI.2390-08.2008 127:547 558. https://doi.org/10.1080/00207454.2016.1209754 259. Díez-Planelles C, Sánchez-Lozano P, Crespo MC et al (2016) 241. Nadim WD, Simion V, Benedetti H et al (2017) MicroRNAs in Circulating microRNAs in Huntington’s disease: emerging medi- neurocognitive dysfunctions: new molecular targets for pharma- ators in metabolic impairment. Pharmacol Res 108:102–110. – cological treatments? Curr Neuropharmacol 15:260 275 https://doi.org/10.1016/j.phrs.2016.05.005 242. Quinlan S, Kenny A, Medina M et al (2017) MicroRNAs in neu- 260. Das E, Jana NR, Bhattacharyya NP (2015) Delayed cell cycle – rodegenerative diseases. Int Rev Cell Mol Biol 334:309 343. progression in STHdh(Q111)/Hdh(Q111) cells, a cell model for https://doi.org/10.1016/bs.ircmb.2017.04.002 Huntington’s disease mediated by microRNA-19a, microRNA- 243. Sethi P, Lukiw WJ (2009) Micro-RNA abundance and stability in 146a and microRNA-432. MicroRNA (Shariqah, United Arab human brain: specific alterations in Alzheimer’s disease temporal Emirates) 4:86–100 lobe neocortex. Neurosci Lett 459:100–104. https://doi.org/10. 261. Reynolds RH, Petersen MH, Willert CWet al (2018) Perturbations 1016/j.neulet.2009.04.052 in the p53/miR-34a/SIRT1 pathway in the R6/2 Huntington’sdis- 244. Li YY, Cui JG, Dua P et al (2011) Differential expression of ease model. Mol Cell Neurosci 88:118–129. https://doi.org/10. miRNA-146a-regulated inflammatory genes in human primary 1016/j.mcn.2017.12.009 neural, astroglial and microglial cells. Neurosci Lett 499:109– 262. Langfelder P,Gao F, Wang N et al (2018) MicroRNA signatures of 113. https://doi.org/10.1016/j.neulet.2011.05.044 endogenous huntingtin CAG repeat expansion in mice. PLoS One 245. Sarkar S, Jun S, Rellick S et al (2016) Expression of microRNA-34a 13:e0190550. https://doi.org/10.1371/journal.pone.0190550 in Alzheimer’s disease brain targets genes linked to synaptic plas- 263. Hoss AG, Lagomarsino VN, Frank S et al (2015) Study of plasma- ticity, energy metabolism, and resting state network activity. Brain derived miRNAs mimic differences in Huntington’s disease brain. Res 1646:139–151. https://doi.org/10.1016/j.brainres.2016.05.026 Mov Disord 30:1961–1964. https://doi.org/10.1002/mds.26457 246. Aloia AL, Abraham AM, Bonder CS et al (2015) Dengue virus- 264. Reed ER, Latourelle JC, Bockholt JH et al (2018) MicroRNAs in induced inflammation of the endothelium and the potential roles of CSF as prodromal biomarkers for Huntington disease in the sphingosine kinase-1 and microRNAs. Mediat Inflamm 2015: PREDICT-HD study. Neurology 90:e264–e272. https://doi.org/ 509306. https://doi.org/10.1155/2015/509306 10.1212/WNL.0000000000004844 Mol Neurobiol (2019) 56:5436–5455 5455

265. Di Pietro L, Lattanzi W, Bernardini C (2018) Skeletal muscle lateral sclerosis using next-generation sequencing. Front Genet 9: microRNAs as key players in the pathogenesis of amyotrophic 310. https://doi.org/10.3389/fgene.2018.00310 lateral sclerosis. Int J Mol Sci 19:1534. https://doi.org/10.3390/ 268. Gershoni-Emek N, Altman T, Ionescu A et al (2018) Localization ijms19051534 of RNAi machinery to axonal branch points and growth cones is 266. Liguori M, Nuzziello N, Introna A et al (2018) Dysregulation of facilitated by mitochondria and is disrupted in ALS. Front Mol MicroRNAs and target genes networks in peripheral blood of pa- Neurosci 11:311. https://doi.org/10.3389/fnmol.2018.00311 tients with sporadic amyotrophic lateral sclerosis. Front Mol 269. Hoye ML, Regan MR, Jensen LA et al (2018) Motor neuron- Neurosci 11:288. https://doi.org/10.3389/fnmol.2018.00288 derived microRNAs cause astrocyte dysfunction in amyotrophic 267. De Felice B, Manfellotto F, Fiorentino G et al (2018) Wide- lateral sclerosis. Brain 141:2561–2575. https://doi.org/10.1093/ ranging analysis of microRNA profiles in sporadic amyotrophic brain/awy182